Granular Flows Down Inclined Rotating Chutes PDF

Granular flows down inclined rotating chutes :
experimental and simulation studies

Citation for published version (APA):
Shirsath, S. S. (2015). Granular flows down inclined rotating chutes : experimental and simulation studies
Eindhoven: Technische Universiteit Eindhoven
Document status and date:

Published: 01/01/2015
Document Version:
Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)
Please check the document version of this publication:
• A submitted manuscript is the version of the article upon submission and before peer-review. There can be
important differences between the submitted version and the official published version of record. People
interested in the research are advised to contact the author for the final version of the publication, or visit the
DOI to the publisher's website.
• The final author version and the galley proof are versions of the publication after peer review.
• The final published version features the final layout of the paper including the volume, issue and page
numbers.
Link to publication
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal.
If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please
follow below link for the End User Agreement:
www.tue.nl/taverne
Take down policy

If you believe that this document breaches copyright please contact us at:
openaccess@tue.nl
providing details and we will investigate your claim.
Download date: 07. Apr. 2019

MGranularMFlowsMDownMInclinedMRotatingMChutesSSMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMSushilMS.MShirsathMMMM
Invitation
toCtheCpublicCdefenseCofCtheC Granular9Flows9Down9
Inclined9Rotating9Chutes
Dissertation
GranularMFlowsMDownM
InclinedMRotatingM Experimental9and9Simulation9Studies9
Chutes
ExperimentalCCandC
SimulationCStudies
onC23rdCofCFebruaryC2715
atC16:77ChrsCinCCollegezaalC4
ofCAuditoriumCofCEindhovenC
UniversityCofCTechnology
AfterCtheCdefense)CthereCwillCbeC
aCreceptionCinCtheCAuditoriumC
SenaatzaalC,south.sidew
SushilMS.MShirsath
sushil@shirsath@gmail@com
Tel:C76.85716243
Paranymphs:
VinayMMahajan
76.86268138
YaliMTang
ISBN:9978-90-386-3786-0 Sushil9S.9Shirsath
76.47611886
Granular Flows Down Inclined Rotating Chutes
Experimental and Simulation Studies
PROEFSCHRIFT
ter verkrijging van de graad van doctor aan de

Technische Universiteit Eindhoven, op gezag van de
rector magnificus, prof.dr.ir. C.J. van Duijn, voor een
commissie aangewezen door het College voor
Promoties in het openbaar te verdedigen
op maandag 23 februari 2015 om 16:00 uur
door
Sushil Shamrao Shirsath
geboren te Jamner, Maharashtra, India

Dit proefschrift is goedgekeurd door de promotoren en de samenstelling van de promotiecom-
missie is als volgt:
voorzitter: prof.dr.ir. J.C. Schouten

promotor: prof.dr.ir. J.A.M. Kuipers
promotor: prof.dr. H.J.H. Clercx
copromotor: dr.ir. J.T. Padding
leden: prof.dr. rer.-nat. S. Luding Universiteit Twente
prof.dr. R.F. Mudde Delft University of Technology
dr.ir. T.W.J. Peeters Tata Steel, IJmuiden, The Netherlands
prof.dr. J.G.M. Kuerten
To my family, friends and well-wishers.
This dissertation is approved by the promotors:
prof.dr.ir. J.A.M. Kuipers
prof.dr. H.J.H. Clercx
and co-promotor:
dr.ir. J.T. Padding
The research reported in this thesis was financially funded by STW.

Nederlandse titel: Granulaire stromingen over roterende glijbanen: Experimentele en simu-
latie studies
©Sushil S. Shirsath, Eindhoven, The Netherlands, 2015
All rights reserved. No part of the material protected by this copyright notice may be re-
produced or utilized in any form or by any means, electronic or mechanical, including pho-
tocopying, recording or by any information storage and retrieval system, without the prior
permission of the author.
Publisher: Gildeprint, Enschede, The Netherlands.

A catalogue record is available from the Eindhoven University of Technology Library
ISBN:978-90-386-3786-0
Summary
The metallurgy industry is one of the major process industries, and its yearly production is
still increasing. The importance of an efficient steel making process is therefore evident. The
reduction of iron ore to pig iron mostly occurs in blast furnaces. The reactant materials for
this reduction, which basically consist of iron ore, coke, and limestone, are charged to the
blast furnace from the top. These materials are distributed on the burden surface via an in-
clined rotating chute. The flow of these materials through the chute can be categorized as
granular flow, which can be described as a highly complex flow with both fluid and solid
characteristics. Often, granular flow suffers from particle segregation due to differences in
size or density, a phenomenon that is undesired. Thus, a better understanding of the funda-
mental phenomena, allowing for a better control of the flow behavior of granular materials
on the rotating chute, is very important for efficient operation of a blast furnace.
To achieve a better understanding of the behaviour of granular flows in rotating chutes, we
initiated a combined experimental and simulation study. A lab-scale set-up of a rotating in-
clined chute was fabricated, where rectangular and semi-cylindrical chute shapes were stud-
ied. The rotation rates were chosen such that they match the characteristics of an industrial
chute according to the Rossby and Froude numbers.
Different measurement techniques were used to measure the surface velocity of the particles
and the particle bed height in the chute. An Electronic Ultrasonic Sensor was used to measure
the bed height. Particle Image Velocimetry (PIV) to measure the two dimensional surface
particle velocity. Finally, Particle Tracking Velocimetry (PTV) has been used to obtain the
surface particle velocity in all three directions (streamwise, spanwise and depthwise), which
is impossible to measure by PIV.
We demonstrated that the PTV technique can also be used to obtain the particle bed height in
the chute. The advantage of the PTV technique is therefore that in one experiment both the
surface particle velocity and the particle bed height can be obtained, avoiding the large num-
ber of experiments needed for bed height measurement using the electronic ultrasonic sensor
v
vi
technique. The experiments showed that the granular flow through an inclined chute is dra-
matically influenced by rotation. An interplay between the Coriolis force and the centrifugal
force is observed during the flow through the chute. The Coriolis force pushes the granular
material to the side wall, causing the flow to decelerate, whereas the centrifugal force causes
the granular flow to accelerate in the streamwise direction.
A Discrete Element Method (DEM) was used to obtain a deeper insight in the granular flow
phenomena. The DEM model was modified by including Coriolis and centrifugal forces into
Newton’s second law of motion for the particle phase and into the Navier-Stokes equations
for the gas phase for a rotating system. The DEM simulations showed that, for the systems
studied here, the interstitial gas has relatively small influence on the granular flow dynam-
ics. A simulation parametric sensitivity study was performed from which we found that the
coefficient of friction plays a dominant role, while the coefficient of restitution and rolling
friction have negligible effects on the flow behaviour. The effect of the precise way of inflow
charging was found to diminish after streaming for 0.2 m along the length of the chute. The
DEM simulation results show a high degree of agreement with the experimental findings for
bed height and surface velocity of particles. After validation, the model was further used
to study phenomena and properties which are difficult to determine experimentally, such as
solid volume fraction and shear velocity flow profile. The DEM model was then extended to
include a rough base consisting of fixed particles with a diameter smaller, equal or larger than
that of the flowing particles. The effect of base roughness on the flow behaviour of monodis-
perse particles was found to be most dominant when the roughness is equal to or larger than
the particle size. Finally, the validated model was used to perform a segregation study of
binary mixture flows of light and heavy particles down non-rotating and rotating chutes. Dif-
ferent cases were studied to understand segregation in the chute as well as segregation in the
radial distribution on the burden surface of a blast furnace. We found that the segregation is
significantly enhanced by chute roughness.
Samenvatting
De metaalindustrie is een van de belangrijkste procesindustrieën, met een nog steeds toene-
mende jaarlijkse productie. Het belang van een efficiënt staalmaakproces is daarom evident.
De reductie van ijzererts naar ruw ijzer vindt voornamelijk plaats in hoogovens. De reac-
terende materialen voor deze reductie, voornamelijk ijzererts, cokes en kalk, worden aan de
bovenkant de hoogoven ingevoerd. De materialen worden door middel van een roterende
glijbaan over het deeltjes oppervlak verspreid. De stroming van deze materialen over de
glijbaan kan gecatagoriseerd worden als een granulaire stroming, welke beschreven kan wor-
den als een hoog-complexe stroming met zowel vloeistof- als vaste stof eigenschappen. Vaak
hebben granulaire stromingen last van deeltjesscheiding ten gevolge van verschillen in grootte
of dichtheid, een ongewenst fenomeen. Een beter begrip van de fundamentele fenomenen,
welke kan leiden tot een betere controle van het stromingsgedrag van granulaire materialen
op roterende glijbanen, is daarom belangrijk voor efficiënte bediening van een hoogoven.
Om een beter begrip te verkrijgen van het gedrag van granulaire stromingen in roterende gli-
jbanen, zijn we begonnen aan een gecombineerde experimentele- en simulatie-studie. Een
laboratorium-schaal opstelling van een roterende glijbaan is in elkaar gezet, waarbij een
rechhoekige en halfronde glijbaan doorsnede zijn onderzocht. De rotatiesnelheden zijn zo
gekozen dat ze overeenkomen met de karakteristieken van een industriële glijbaan volgens
de Rossby en Froude kentallen.
Verschillende meettechnieken zijn gebruikt om de oppervlaktesnelheid van de deeltjes en de
hoogte van het deeltjesbed in de glijbaan te meten. Een electronische ultrasone sensor is
gebruikt om de hoogte van het deeltjesbed te meten. Particle Image Velocimetry (PIV) is
gebruikt om de tweedimensionale oppervlaktesnelheid te meten. Tenslotte is Particle Track-
ing Velocimetry (PTV) gebruikt om de oppervlaktesnelheid in alle drie richtingen (lengte,
breedte en diepte) te meten, wat onmogelijk is via PIV.
We hebben gedemonstreerd dat de PTV techniek ook gebruikt kan worden om de bedhoogte
in de glijbaan te verkrijgen. Het voordeel van de PTV techniek is daarom dat in één exper-
vii
viii
iment zowel de oppervlaktesnelheid en de bedhoogte verkregen kunnen worden, waarmee

het grote aantal experimenten benodigd voor bedhoogte metingen met de electronische ultra-
sonische sensor voorkomen kan worden. De experimenten hebben aangetonnd dat de gran-
ulaire stroming door een glijbaan dramatisch wordt beïnvloed door rotatie. Er treedt een
samenspel tussen de Corioliskracht en centrifugaalkracht op gedurende de stroming door de
glijbaan. De Corioliskracht duwt het granulaire materiaal naar de zijwand, waardoor de stro-
mingssnelheid afneemt, terwijl de centrifugaalkracht de granulaire stroming versnelt in de
stromingsrichting.
Een Discrete Element Methode (DEM) is gebruikt om een dieper inzicht te verkrijgen in
de granulaire stromingsfenomenen. Het DEM model is aangepast door Coriolis- en cen-
trifugaalkrachten mee te nemen in Newton’s tweede hoofdwet voor de deeltjesfase en in de
Navier-Stokes vergelijkingen voor de gasfase voor een roterend systeem. De DEM simulaties
hebben aangetoond dat, voor de systemen die hier onderzocht zijn, het gas een relatief kleine
rol speelt in het stromingsgedrag. Een simulatie parameter gevoeligheidsstudie is uitgevoerd,
waaruit we hebben geleerd dat de frictiecoëfficiënt een dominante rol speelt, terwijl de resti-
tutiecoëfficiënt en rolfrictie een klein effect hebben op het stromingsgedrag. Het effect van
de precieze manier waarop de deeltjes worden ingebracht in de glijbaan raakt verloren na een
stroming van 0.2 m in de glijbaan. De DEM simulaties laten een grote overeenkomst zien
met de experimentele bevindingen voor deeltjes bedhoogte en snelheid aan het oppervlak. Na
validatie is het model verder gebruikt om fenomenen en eigenschappen te bestuderen welke
moeilijk experimenteel bepaald kunnen worden, zoals de deeltjes volumefractie en afschuif-
snelheidsprofiel. Het DEM model is daarna uitgebreid door een ruwe bodem van gefixeerde
deeltjes aan te brengen, met een deeltjesdiameter kleiner, gelijk aan, of groter dan die van de
stromende deeltjes. Het effect van bodemruwheid op het stromingsgedrag van monodisperse
deeltjes bleek het meest dominant te zijn wanneer de ruwheid gelijk aan of groter dan de
deeltjesgrootte is. Tenslotte is het gevalideerde model gebruikt om een segregatiestudie uit
te voeren van een binair mengsel van lichte en zware deeltjes in niet-roterende en roterende
glijbanen. Verschillende gevallen zijn onderzocht om de segregatie in de glijbaan te begri-
jpen, alsook segregatie in de radiële distributie op het deeltjesoppervlak van een hoogoven.
De segregatie bleek sterk toe te nemen met toenemende ruwheid van de glijbaan.
Table of contents
1 Introduction 1
1.1 General introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Granular material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 Blast furnace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Charging mechanism in the blast furnace . . . . . . . . . . . . . . . . . . . . 4
1.5 Objective of the thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.6 Experimental approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.7 Discrete Element Method simulations . . . . . . . . . . . . . . . . . . . . . 6
1.8 Outline of the thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Experimental study of granular flow down a rotating chute 9

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Experimental setup and techniques . . . . . . . . . . . . . . . . . . . . . . . 12
Experimental setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Dynamic weighing scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Ultrasonic height sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Particle image velocimetry . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Experimental procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Experimental settings and materials . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Results and discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Mass flow rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Particle bed height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Surface particle velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3 Numerical investigation of granular flow down a rotating chute 25

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
ix
x
3.2 Simulation model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Equations of motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Soft-sphere contact model . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Coriolis and centrifugal forces . . . . . . . . . . . . . . . . . . . . . . . . . 32
Equations for gas hydrodynamics . . . . . . . . . . . . . . . . . . . . . . . . 33
Simulation settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.3 Computational measurements . . . . . . . . . . . . . . . . . . . . . . . . . 36
Calculation of bed height in simulations . . . . . . . . . . . . . . . . . . . . 36
Calculation of surface velocity in simulations . . . . . . . . . . . . . . . . . 36
3.4 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Influence of inlet particle charging . . . . . . . . . . . . . . . . . . . . . . . 38
Influence of interstitial gas . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Sensitivity study of collision parameters . . . . . . . . . . . . . . . . . . . . 40
3.5 Validation: comparison with experimental results . . . . . . . . . . . . . . . 45
Particle bed height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Streamwise surface velocity in the width-wise direction . . . . . . . . . . . . 47
Spanwise surface velocity in the width-wise direction . . . . . . . . . . . . . 50
3.6 Computational measurements of optically inaccessible properties . . . . . . . 50
Depth dependent flow profiles . . . . . . . . . . . . . . . . . . . . . . . . . 50
Depth dependent solid volume fraction . . . . . . . . . . . . . . . . . . . . . 51
3.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4 Cross-validation of 3D-PTV for granular flows down rotating chutes 55

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2 Experimental setup and procedure . . . . . . . . . . . . . . . . . . . . . . . 58
Experimental setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Experimental procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3 Particle tracking velocimetry technique . . . . . . . . . . . . . . . . . . . . 63
Calibration of the camera system . . . . . . . . . . . . . . . . . . . . . . . . 64
Three-dimensional positioning of the particles . . . . . . . . . . . . . . . . . 64
Particle trajectory reconstruction . . . . . . . . . . . . . . . . . . . . . . . . 65
4.5 Postprocessing of experimental and simulation data . . . . . . . . . . . . . . 67
Calculation of bed height from experimental PTV data . . . . . . . . . . . . 67
Calculation of bed height in simulations . . . . . . . . . . . . . . . . . . . . 67
Calculation of experimental surface velocity using PIV . . . . . . . . . . . . 69
Calculation of experimental surface velocity using PTV . . . . . . . . . . . . 69
xi
Calculation of surface velocity in simulations . . . . . . . . . . . . . . . . . 69

4.6 Experimental results and discussion . . . . . . . . . . . . . . . . . . . . . . 70
Repeatability of PTV experiments . . . . . . . . . . . . . . . . . . . . . . . 70
Comparison of bed height using PTV and the ultrasonic sensor . . . . . . . . 70
Comparison of streamwise velocity using PTV and PIV . . . . . . . . . . . . 73
4.7 Validation of DEM using experimental measurements . . . . . . . . . . . . . 73
Bed surface height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Streamwise surface particle velocity . . . . . . . . . . . . . . . . . . . . . . 74
Spanwise surface particle velocity . . . . . . . . . . . . . . . . . . . . . . . 75
Depthwise surface particle velocity . . . . . . . . . . . . . . . . . . . . . . . 77
Depth dependent solid volume fraction . . . . . . . . . . . . . . . . . . . . . 78
4.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5 Effect of wall roughness on granular flows down rotating semi-cylindrical chutes 81

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.2 3D PTV measurements for rotating granular flows . . . . . . . . . . . . . . . 86
Discrete element model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Implementation of the smooth and rough wall chute . . . . . . . . . . . . . . 86
Simulation and experimental settings . . . . . . . . . . . . . . . . . . . . . . 87
Computational measurements . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.4 Validation of the discrete element model for a smooth semicylindrical chute . 91
Streamwise surface particle velocity as a function of width position . . . . . . 91
Spanwise surface particle velocity as a function of width position . . . . . . . 93
Bed height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.5 Influence of base roughness on monodisperse granular flows . . . . . . . . . 95
Bed height along the length of the chute . . . . . . . . . . . . . . . . . . . . 95
Streamwise particle velocity as a function of the position along the chute . . . 97
Streamwise particle velocity as a function of width position . . . . . . . . . . 98
Streamwise velocity as a function of height in the chute (velocity shear profiles) 99
Granular temperature as a function of the position along the chute . . . . . . 100
Distribution and anisotropy of granular temperature . . . . . . . . . . . . . . 101
5.6 Influence of base roughness on segregation in binary density flows . . . . . . 106
Center of mass height along the length of the chute . . . . . . . . . . . . . . 106
Granular temperature and equipartition of energy . . . . . . . . . . . . . . . 109
5.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
xii
6 Simulation study of density segregation down rotating rectangular chutes 111

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Granular flow down an inclined rotating chute . . . . . . . . . . . . . . . . . 114
Trajectories from the chute exit to the burden surface . . . . . . . . . . . . . 114
Simulation settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
6.3 Overview of cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
6.4 Case 1: effect of the chute inclination angle . . . . . . . . . . . . . . . . . . 118
Center of mass height and segregation rate . . . . . . . . . . . . . . . . . . . 118
Radial burden distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
6.5 Case 2: effect of particle feeding mass rate . . . . . . . . . . . . . . . . . . . 121
6.6 Case 3: effect of density ratio of the particles . . . . . . . . . . . . . . . . . 123
6.7 Case 4: effect of chute roughness and insertion of a dividing wall in the chute 125
6.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
7 General conclusions and recommendations 133

7.1 General conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
7.2 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Different experimental and simulation approaches . . . . . . . . . . . . . . . 135
A Appendix: Single particle experiments 139

A.1 Coefficient of restitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
A.2 Coefficient of friction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
References 143
Nomenclature 159
xiii
Publications 163
Acknowledgments 165
About the author 167

Chapter
1
Introduction
1.1 General introduction

The blast furnace, which was invented around the first century AD, is still one of the major
methods to produce steel from iron ore. Nowadays, the yearly steel production is around 1500
million tonne world wide, and this number is still increasing every year. Mostly, the steel is
produced in blast furnaces and thus the efficiency of the blast furnace is of great importance.
The modern iron blast furnace is the result of centuries of development. The development
of the blast furnace through the years is aided by an increase in understanding of the steel
making process (Rosenqvist, 2004). This process of understanding is still ongoing, and con-
sequently the efficiency of the steel making process can be increased. This thesis focuses
on one aspect of the steel making process, namely the charging (feeding mechanism) of the
granular material into the blast furnace reactor. The charging of the granular material occurs
via a feeder at the top of the blast furnace, more specific a rotating inclined chute. Before
going into more detail, first the main principles and characteristics of granular materials, blast
furnace, and rotating inclined chute (blast furnace feeder) are introduced. Secondly, the thesis
objectives and the experimental and simulation approach required to fulfill the objectives are
discussed briefly. Finally, the outline of this thesis is presented.
1.2 Granular material

Flows of granular material or avalanches are often seen in nature as well as in industry. Ap-
proximately one-half of the products, and more than three-quarters of the raw materials used
in chemical industries, are in granular form (Nedderman, 2005). Granular materials behave
sometimes like a solid, sometimes like a liquid, but most often differently from both, they
are sometimes simple other times complex, sometimes predictable other times surprising;
their behaviour is extremely diverse and rich (Jaeger et al., 1992, 1996). Granular material
can be defined as individual solid particles composed of any material, with any particle size
1
2 Chapter 1
Figure 1.1: Raw granular material used in the blast furnace process. The approximate value
for bulk density of coke is 900 kg/m3 and size is 25-40 mm, density of pellets 4000 kg/m3
and size is 10-15 mm, and density of sinter is 2000 kg/m3 and size 10-25 mm.
or particle density (Nedderman, 2005). Mostly, the diameter, shape, and density are irregu-
lar within a certain granular flow. Non-uniformity in the material feed flow generally leads
to particle segregation and consequently to a decrease in product quality and discontinuous
operation. Obviously, a well-mixed granular flow is therefore desirable. Therefore, funda-
mental understanding of particle segregation in granular flows is essential to determine the
success or failure of a vast number of man-made and natural processes. Heuristics are com-
monly used in order to limit the degree of segregation in flows. However, these heuristics
are often unreliable for scale-up and too specific (Michaels, 2003), leading to suboptimal
performance for particulate process operations (Wibowo and Ng, 2001). A more preferable
approach is a quantitative prediction of the flow behavior to gain insight into the segregation
of the particles, dependent on the flow conditions and physical properties of the material.
Typical applications of granular flow can be found in operations in chemical, petroleum, food,
pharmaceutical, electronic, biochemical, mining, and metallurgical industry (Sun, 2007).
Many solids processing devices such as mixers, chutes, hoppers, and other transfer equip-
ment are used in industry to handle granular materials. The understanding of the flow of
granular materials is important for successful operation for example, transporting, mixing,
heating, granulation, blending, pressing, and drying etc. Generally, these operations could
reap the benefits from a quantitative approach, as the efficiency can be improved by under-
standing and predicting the granular flow behavior in a quantitative manner. In this research,
such an approach is used, inspired by granular flows occurring in the metallurgical industry.
To be more specific, the work described in this thesis is relevant for understanding granular
flows occurring within the charging process of a blast furnace. Figure 1.1 shows the typical
raw granular materials used in blast furnaces, namely coke, iron-ore, and pellets. They show
a clear diversity in shape and size. Moreover, the density variation of the granules plays an
important role in these applications.
Introduction 3
1.3 Blast furnace
Figure 1.2: Schematic of a blast furnace used for the production of steel. The coke and iron
ore is charged from the top hopper into the blast furnace through an inclined rotating chute
with typical rotation rates of 8-10 rpm. Alternate layers of coke and iron ore are charged on
the burden surface. The molten slag and molten iron is collected from the bottom.
A typical blast furnace reactor has more than 5000 m3 in volume used to produce pig-iron
from iron-ore particles. A schematic representation of the blast furnace is shown in figure
1.2. The raw material, consisting of iron ore, coke particles, and limestone, is charged at the
top of the blast furnace. Firstly, the raw materials are hoisted and deposited into the hopper
which is situated above the chute. The raw material consists of particles with different size,
density and shapes, as shown in the figure 1.1. The distribution of raw materials at the burden
surface is very crucial for the smooth operation of the blast furnace. This is achieved with
the use of an inclined rotating chute shown in figure 1.3a, whose rotation rate and inclination
angle can be flexible and adjusted as needed. The raw material is charged in a batch-wise
4 Chapter 1
process to the reactor. The raw material is deposited in the blast furnace in alternating layers
of iron ore and coke, and then gas is blown from the tuyeres at the bottom of the reactor at
high temperature and pressure. Iron ore particles are reduced to pig-iron while descending,
and many physical changes and complete chemical reactions occur between species residing
in different phases. The pig-iron is further treated in order to lower the carbon content and to
produce steel. Few ways in which the blast furnace efficiency can be controlled is by varying
the quality of the raw materials, the burden distribution and the air blasting condition. The
burden distribution affects the operating efficiency strongly (Mio et al., 2009). Since burden
distribution depends on the burden descending behavior from the rotating chute, any particle
segregation during charging at the top of the blast furnace affects its performance to a large
degree. Non-uniformity in the material feed leads to segregation in granular flows. Particle
segregation decreases product quality and affects operation. A well-mixed granular flow is
therefore desirable. This makes a fundamental understanding of flow behaviour of granular
material on a rotating chute essential.
1.4 Charging mechanism in the blast furnace

Generally, two types of blast furnace charging methods are used in the industry: the first one
is a bell-less charging system shown a schematic representation of figure 1.3a and the other
is a bell-type charging system shown a schematic representation of figure 1.3b. Traditionally,
the bell-type charging system was used in blast furnaces (with an effective volume of 1000 m3
or less). However, since the invention of the bell-less charging system in the early 70s, this
type is commonly used nowadays in newer blast furnaces. In contrast to the bell-type system,
the bell-less system has the main advantage that this type has more flexibility in charging
and therefore it is possible to distribute the burden more efficiently see, Yu (2013). The
rotating chute provides the bell-less charging system with the ability to vary the inclination
angle, from which the system gains its flexibility to operate in different charging patterns.
Although such a rotating chute is commonly used in industry, a fundamental understanding of
the particle behavior and segregation through a rotating inclined chute is lacking in literature.
Key is to control the feed of the granular matter to the blast furnace. Therefore, we need to
understand the flow dynamics and predict the granular flow in the chute.
1.5 Objective of the thesis

The main objective of this thesis is to validate a simulation model tool to study the dynamics
of monodisperse and bidisperse granular flows down inclined rotating chutes. To achieve
this, the first task is to build a rotating chute flow setup and perform high-quality experiments
for different angles, rotation rates and flow rates. Different measurement techniques are em-
ployed such as an ultrasonic sensor for particle bed height, particle image velocimetry (PIV)
Introduction 5
Recieving hopper
Iron ore Coke
Rotating chute
(a)
(b)
Figure 1.3: Schematic of (a) bell-less and (b) bell-type charging mechanism of a blast furnace.
6 Chapter 1
for surface velocity of particles and particle tracking velocimetry (PTV) to determine the sur-
face velocity particle as well as bed height in a single experiment. The surface streamwise
and spanwise velocities of the particles and the particle bed height are used for validation of
the DEM simulation model. The validated model can then be used to study segregation in
chutes, including chutes with a rough base, and to study segregation for different operating
conditions of the chute.
1.6 Experimental approach

For the models developed to be reliable, it is essential that they are subjected to a validation
process in which the models are compared with real systems. Generally, models are validated
by comparing qualitative and quantitative measurement of macroscopic parameters with the
corresponding data obtained from experiments. The developed rotating chute flow setup is
designed in a such way that the dimensionless Rossby number and Froude number are com-
parable to those of the industrial chute under normal operating conditions. The experiments
were performed on the rotating table of the Department of Applied Physics at TU/e. Ex-
perimental measurement techniques can be divided into two categories. The first category
concerns intrusive techniques in which the measurement technique interferes with the flow
dynamics of granular materials. The second category consists of non-intrusive techniques
which do not interfere with the flow. In the research work described in this thesis, only non-
intrusive measurement techniques are used, with the disadvantage that they can only observe
surface phenomena of granular flows. As we will show, these observations are for our purpose
sufficient to validate the simulation model.
The non-intrusive measurement techniques adopted in this work are Particle Image Velocime-
try (PIV) to measure the surface velocity of particles in both streamwise and spanwise direc-
tion in the rotating chute. An Electronic Ultrasonic Sensor is used to measure the particle bed
height in the rotating chute at different locations in the chute as the flow is not uniform in the
streamwise and spanwise directions. Three-dimensional Particle Tracking Velocimetry (3D-
PTV) is applied to measure the streamwise, spanwise and depthwise velocity components of
tracer particles in the chute, as well as the particle bed height in the chute. These experiments
are conducted to gain insight into the flow phenomena that occur in the rotating chutes and to
compare for validation purposes experimental results with the simulation data obtained from
the Discrete Element Method (DEM).
1.7 Discrete Element Method simulations

The Discrete Element Method (DEM) is one of the key simulation tools to study the flow
behaviour of granular materials. It is also providing insight into the mechanisms governing
Introduction 7
particulate flow and it is a powerful tool for optimising a number of industrial processes. Such
numerical simulations can enhance fundamental understanding of granular motion and can
also help in the improvement of design or operation of systems involving particulate materi-
als (Cleary, 2010). The most popular version of DEM is based on the linear spring-dashpot
model introduced by Cundall and Strack (1979) to study rock dynamics, but it has found
wide applications in other fields involving granular materials. Newton’s laws of motion form
the theoretical basis for the DEM. In the DEM, the total force acting on individual grains or
particles in a granular medium is evaluated and the accelerations, velocities and positions that
result from this force are tracked over a period of time. DEM is clearly a useful tool for sim-
ulation of granular flows in inclined channels, yet is confronted with considerable challenges
in addition to its inherent computational limitations. In the DEM, particles are commonly
represented as perfect spheres for simplicity and computational efficiency. However, stud-
ies have shown that the interlocking of non-spherical particles at arbitrary edges influences
significantly the behavior of granular flow systems. Another concern is the assumption that
the macro patterns in granular media can be represented by various input parameters when
the model is calibrated. Furthermore, a specific combination of parameters might not work
for all processes. Yet another concern is the validation of model results. Generally, model
results are compared with experiments in terms of qualitative visualizations of flow patterns
and quantitative representations of macroscopic features. In this approach, the sensitivity
of the model results to input parameters like the coefficient of friction, rolling friction, and
restitution is examined.
In this research, we analyse and study the system from a co-rotating frame of reference. The
resulting Coriolis and centrifugal forces are included in Newton’s law equations to study the
effect of rotation on the flow behavior of granular material. These forces are also included into
the gas phase equations in order to check the influence of interstitial gas on the flow dynamics
in both rotating and non-rotating chutes. A semi-cylindrical chute is also implemented with
either a flat frictional base or a bumpy base with varying base roughnesses. The segregation
of a binary mixture of spherical particles of high and low density is studied in the DEM. The
trajectories of particles leaving the chute, hitting the burden plane, is calculated theoretically
in the model.
1.8 Outline of the thesis

This thesis contains a detailed report of the activities that have been embarked upon in the
utilization and modification of the discrete element model; a comprehensive report on the
experiments that have been conducted for the validation of the DEM model, and the obtained
understanding of granular flow behaviour in general. The results obtained from the studies
have been analyzed critically, and have lead to several main conclusions.
8 Chapter 1
In detail, chapter 2 presents an experimental study of monodisperse flows through inclined

rotating chutes. The effect of rotation of the chute on the flow behaviour of monodisperse
particles of 3 mm in the chute is studied. Particle Image Velocimetry (PIV) and Electronic
Ultrasonic sensor techniques are used to measure the surface particle velocimetry and bed
height in the chute, respectively.
In Chapter 3, Discrete Element Method (DEM) simulations are performed for monodisperse
flows through inclined rotating chutes. The DEM simulation results are validated by compar-
ing with the experimental results of chapter 2. The influence of gas on the flow behaviour is
also studied under both non-rotating and rotating conditions. The effect of simulation param-
eters as well as inflow charging conditions on the flow behaviour is also studied.
In Chapter 4, the Particle Tracking Velocimetry (PTV) technique is applied to granular flows
through inclined rotating chutes. This PTV technique is cross-validated by the PIV technique
for surface particle velocity and against the electronic ultrasonic sensor technique for bed
height measurement. The feeding of particles at the inlet, implemented in the DEM model,
is validated using the PTV technique for higher mass flow rates too.
In Chapter 5, the rectangular chute in the DEM model is modified to a semicylindrical chute
shape and then validated using the PTV techniques for monodisperse granular flows. Further,
the influence of the roughness of the base of the chute on the flow behaviour of monodisperse
particles is studied. A density segregation study is is also performed for different rotation
rates and varying base roughnesses using DEM.
In Chapter 6, the validated DEM model is used to study density segregation in the inclined
rotating rectangular chute. The radial deposition distribution of low and high density particles
is calculated from the DEM model. The effect of inclination angle, mass flow rate, density
ratio, base roughness, chute wall roughness and insertion of another wall into a rough chute
is studied.
In Chapter 7, we give our conclusions and recommendations.
Chapter
2
Experimental study of granular flow
down a rotating chute ∗
In blast furnaces, particles like coke, sinter and pellets enter from a hopper and are dis-
tributed on the burden surface by a rotating chute. Such particulate flows suffer occasionally
from particle segregation during transportation caused by differences in density or size. To
get a more fundamental insight into these effects, we started an experimental study to inves-
tigate the effect of rotation on such particulate flows. Here, as a first step, we present an
experimental study of granular flow of monodisperse 3 mm spherical glass particles flowing
with a constant mass rate through a rotating smooth rectangular chute, which is inclined at
a fixed angle. Experiments are performed for a sufficiently long time to study steady (but
streamwise accelerating) flow. Particle image velocimetry (PIV), electronic ultrasonic height
sensors, and a dynamic weighing scale are used to measure the surface velocity of the particle
stream, the particle bed height and mass flow rate in the chute, respectively. The influence of
rotation speed and angle of inclination of the chute is studied. We find an interesting interplay
between the Coriolis force, which pushes the granular flow to the sidewall of the chute and
tends to diminish the acceleration of the flow, and centrifugal forces that accelerate the flow.
The velocity components display a complex dependence on the spanwise and streamwise po-
sition in the chute. The bed height in the chute is also influenced by these effects of system
rotation. These measurements provide a well-defined set of observations for refining and val-
idating computer simulations of granular flows, and point out some important limitations of
physical experiments.
∗ This chapter is based on: Shirsath, S.S., Padding, J.T., Deen, N.G., Clercx, H.J.H. and Kuipers, J.A.M. (2013).
Experimental study of monodisperse granular flow through an inclined rotating chute. Powder Technology, 246,
235-246
9
10 Chapter 2
2.1 Introduction
The flow behaviour of dry granular materials is very important in many man made and nat-
urally occurring processes. Chemical, pharmaceutical, metallurgical and agricultural indus-
tries need to handle granular materials to produce their final products. Handling of granular
materials involves the application of many solids processing devices such as mixers, chutes,
hoppers, and other transfer equipment. An important example in the steel industry concerns
the controlled charging of blast furnaces with coke, sinter and pellets, where the particles are
finally delivered through open rotating channels. Despite their common occurrence, funda-
mental knowledge about the dynamics of particulate flows through inclined open channels
is still limited. Such particulate flows suffer occasionally from particle segregation during
transportation caused by differences in density or size. The behaviour of a granular flow
through an inclined channel is partly determined by the wall surface characteristics, which
can be either smooth or rough. Smooth surfaces lead to an ever-accelerating flow (Augen-
stein and Hogg, 1978; Brennen et al., 1983; Campbell and Brennen, 1985b; Johnson et al.,
1990; Holyoake, 2011). In case of a rough surface, for a range of chute angles a slow steady
uniform granular flow can be produced. Such a rough surface is often generated by gluing
particles to the surface (Pouliquen, 1999; MiDia, 2004; Holyoake, 2011; Sheng, 2012). It is
of interest to mention that GDR Midi issues a phenomenological constitutive law for flows
over uniform but rough chutes. In contrast to classical fluid mechanics, general constitutive
laws for granular flows are largely unknown and experimental studies in these circumstances
provide a key approach to the investigation of their dynamical behavior. Several experimental
investigations and simulations have been carried out in the past on complex conveyer geome-
try and profiles to study the bulk flow behavior of granular materials (Roberts, 1969; Roberts
and Scott, 1981). Continuum-based methods have been used to predict the flow of granular
cohesionless materials in chutes (Roberts, 2003). Different measurement techniques are used
for measurement of granular flow based on force sensors (Savage and McKeown, 1983; Zenit
et al., 1997), acoustic probes (Bennett and Best, 1995) , tracked transmitters (Dave et al.,
1999), capacitance probes (Louge et al., 1997), optical sensors (Dent et al., 1998), and digital
imaging (Guler et al., 1999; Capart et al., 2002; Bonamy et al., 2002). Variations on uniform
granular flow pertaining to flow around obstacles and oblique granular jumps or shock waves
at slight corners have also been studied (Gray et al., 2003; Hákonardóttir and Hogg, 2005).
We consider specifically the effect of chute rotation on the granular flow of monodisperse
particles down a smooth inclined chute. The novelty of our investigation lies in the fact that
state-of-the-art experimental methods have been used to obtain a well-defined reference data
set of particles flowing down an inclined chute at different rotation rates. The rotation leads to
additional Coriolis and centrifugal forces on the particles shown in figure 2.1, which in turn
lead to flow paths and particle distributions in a rotating chute which deviate considerably
Experimental study of granular flow down a rotating chute 11
-ma(x(xr))
-2ma(xva)
va
m ag
y
z
Figure 2.1: Orientation of gravitational, centrifugal and Coriolis forces experienced by parti-
cles flowing down a rotating inclined chute.
from those in a non-rotating chute. The key parameter that classifies the relative effect of the
inertial and the Coriolis forces is the so-called Rossby number, defined as
va
Ro = , (2.1)
2ΩL cos φ
where va is the particulate phase velocity (within the co-rotating frame), Ω is the rotation rate
of the chute, L is a chute length and φ is the angle of inclination of the chute with respect to the
horizontal direction. This is an important dimensionless parameter also in geophysical flows
when comparing typical flow time scales vLa with relative rate of the Earth. When Ro 1, the
effects of rotation are unimportant and can be neglected, whereas Ro 1 signifies a system
which is strongly affected by Coriolis forces. The relative importance of the centrifugal force
compared to the gravitational force can be quantified by a (rotational) Froude number defined
as Mellmann (2001),
Ω2 L cos φ
Fr = . (2.2)
g
12 Chapter 2
When Fr 1 gravity dominates the centrifugal effects which keeps all particles inside the
chute.
This chapter is organized as follows. In section 2 we present the experimental set-up and the
measurement techniques. In section 3, we present the experimental results of the effects of
rotation on the flow behavior of a monodisperse granular flow through an inclined chute for
three different inclination angles φ =30, 40 and 50 degrees. We will show that pronounced
effects of rotation occur in granular flows for which Ro < 1. We give our conclusions in
section 5.
2.2 Experimental setup and techniques

In this section we describe the experimental setup and measurement techniques, including
use of electronic ultrasonic height sensors and particle image velocimetry (PIV). We outline
experimental settings and the experimental protocol followed in the experiments.
Experimental setup
In Figure 2.2 we give a schematic representation of the side view of the experimental setup.
The chute is made of plexiglass for optical accessibility. The experimental equipment in-
cludes a hopper for storage of particles and a collection tank. A vacuum pump is attached
to the top of the hopper to recycle the granular material from the collection tank. The mass
rate is measured through a dynamic weighing scale, the particle bed height is measured at
selected locations through an ultrasonic sensor, and the surface particle velocity field is mea-
sured through a PIV camera and basic PIV tools.
The granular material is stored in a hopper at the top end of the chute. To minimise rotational
flow of the granular material inside the hopper prior to deposition on the chute, the rotation
axis passes through the centre of the hopper. The bottom end of the hopper is connected with
different mouthpieces in order to get different flow rates at the exit of the hopper. The flow
region of the experimental setup consists of a rectangular Plexiglas chute with transparent
sidewalls through which the moving granular material can be photographed. The interior
of the chute is 1 m in length, 10 cm in height, and 8 cm in width. The bottom wall of the
chute is black to achieve a better contrast in the photographs. The inclination angle of the
chute (defined with respect to the horizontal) is adjustable between 0 and 70 degrees. At
the end of the chute, a collecting tank is placed on a dynamic weighing scale to measure the
flow rate. The PIV camera is placed halfway the length of the chute, perpendicularly to the
chute at a distance of 1.6 m. A 28 mm lens is used to cover the entire length of the chute
in a single photograph. The whole setup is mounted on a rotating table, so that the flow is
measured in the non-inertial frame of reference. All the hardware constituting the focusing

PIV camera
Hopper
Support stand
Plexiglass chute
Collection tank
Weighing scale
Rotating table
and acquisition
hardware
Figure 2.2: Schematic drawing of the experimental PIV setup, side view.
system and the measurement system is located on the top of the table or surface below it.
The equipment is remotely controlled from an adjacent room for safety precautions during
the rotating experiments. Details of the experimental techniques will be described in the next
subsections.
Dynamic weighing scale

The mass rate with which the granular particles are flowing through the chute in steady state
is measured by a dynamic weighing scale of 0-60 kg capacity provided by Mettler Toledo,
installed at the downstream end of the chute. Care has been taken to place the weighing scale
14 Chapter 2
horizontally to prevent centrifugal forces from influencing the measured weight. Weight mea-
surements are logged at intervals of 1.0 sec. Mass rate information acquired by the weighing
scale is collected and averaged over a time interval during which the flow is steady. The error
in the measurements of the mass rate during rotating experiments is +/- 0.05 kg/s.
Ultrasonic height sensor

We use an electronic ultrasonic sensor LRS3, provided by Formate Messtechnik GmbH (Ger-
many), to measure the particle bed height at selected locations in the chute. The sensor sends
an ultrasonic acoustic wave, then measures the time necessary for the wave to return by re-
flection against a free surface, and finally computes the average distance separating it from
the free surface. We use a circular sensor of 28 mm in diameter, resulting in measurements of
the average height of the particle bed in a circular area of approximately the same diameter.
The ultrasonic sensor is connected to a computer via a data acquisition device. Height infor-
mation acquired by the sensor is collected, filtered, and averaged over a time interval during
which the flow is steady. The error in the height measurements is about 10% of a particle
diameter according to independent experiments The sensor is fixed at different positions in
the chute along the width and length directions.
Particle image velocimetry

Particle Image Velocimetry (PIV) is a non-invasive measuring technique developed origi-
nally to investigate liquid or gas–liquid systems, but recently extended to gas–solid dispersed
flows. This measurement technique can be used to obtain the instantaneous velocity field
(Keane and Adrian, 1991; Westerweel, 1997). The basic principle of PIV is to record two
images with a short time delay and determine the displacement of the particles between the
two images with a spatial cross-correlation algorithm on two consecutive images. Another
method is Particle Tracking Velocimetry (PTV) which uses the displacement of the individ-
ual particles to determine the velocity. An important difference between the PIV and PTV is
the seeding density which is much lower in PTV experiments. In PIV analysis, the images
are divided into NxN interrogation areas and a velocity field can be obtained. More recently,
PIV has also been successfully applied to granular systems (Link et al., 2004; Bokkers et al.,
2004; Santana et al., 2005; Dijkhuizen et al., 2007). In this case no laser is required, since
the particles are now illuminated directly from the front using LED lamps. Since granular
systems are opaque in nature, these measurements are restricted to the study of the particles
visible at the flow surface. When a PIV analysis is initiated, an image is subdivided into small
‘interrogation’ areas of N × N pixels. These interrogation areas do not need to be strictly ad-
jacent to each other. Usually, they overlap, so that a more detailed velocity field is obtained.
For each interrogation area a cross-correlation is performed between two subsequent images,
according to:
1 N N
R(sx , sy ) = 2 ∑ ∑
N i=1 j=1
(I [i, j] − I )(I [i + sx , j + sy ] − I ) (2.3)

for every displacement vector (sx , sy ). I [i, j] is the intensity of pixel [i, j] in the first image and

I [i + sx , j + sy ] is the intensity of pixel [i + sx , j + sy ] in the second image. Fluctuating noise

is eliminated by subtracting the average intensities I and I of the first and second image of
the interrogation area under consideration, respectively.
The instantaneous velocity can now be obtained by dividing the peak displacement s a with
the time delay between the images and the magnification M (in pixels per meter) of the image,
i.e.
s a (xx,t)
v a (xx,t) = . (2.4)
MΔt
Careful selection of the time between two consecutive images is required to minimize the in-
fluence of out-of plane movement of particles. We have set the resolution and magnification
in such a way that the particle images are about 5 pixels in diameter, which is known to give
accurate results in PIV measurements (Westerweel, 1997; Dijkhuizen et al., 2007). A multi-
pass algorithm was used with an initial interrogation area size of 32 pixels and a final size of
16 pixels non-overlapping interrogation areas, which yields an approximated displacement
error of O(0.1) pixels (Westerweel, 1997). Due to the very high seeding density (particle
image pairs per interrogation area) there are practically no outliers. Any remaining outliers
were removed with a standard median filter. After post-processing, a time-averaged velocity
field is obtained from a sequence of instantaneous velocity fields obtained over a time interval
during which the flow is steady.
Setting of camera Images with a resolution of 1080x170 pixels were recorded with a LaV-
ision ImagerPro HS CCD camera which has an internal memory of 2 GB. For the PIV mea-
surements, the camera is located perpendicularly at a distance of 1.6 m from the base of the
chute. Using a 28 mm lens this leads to a measurement area of approximately 75 cm x 8 cm
and 4 to 5 pixels for each particle diameter, which is high enough to obtain the desired spatial
resolution (Westerweel, 1997). The bottom end 1.5 cm of the chute just close to collection
tank could not be studied due to lack of visual accessibility. The frequency with which PIV
image pairs are recorded is 100 Hz. The exposure time is set to 0.3 ms with an effective time
delay of 1.2 ms between the images in a pair. With this scheme, the camera is able to record
for 45 sec.
16 Chapter 2
Table 2.1: Experimental settings and materials for experiments.
Property Value
Length of chute 0.1 m
Width of chute 0.08 m
Height of chute 0.09 m
Inclination of chute 30, 40, 50 deg
Mass flow rate 1.6 kg/s
Particle type Spherical glass
Particle diameter 3 mm
Particle density 2550 kg/m3
Table 2.2: Rossby and Froude numbers based on observed velocities at the end of the chute
for different angles of inclination and rotation rates.
Rossby number Froude number

deg Ω = 4rpm Ω = 8rpm Ω = 16rpm Ω = 4rpm Ω = 8rpm Ω = 16rpm
30 3.68 1.84 0.92 0.013 0.055 0.223
40 5.02 2.51 1.26 0.012 0.049 0.197
50 6.71 3.35 1.68 0.010 0.041 0.165
Experimental procedure
The following steps are used during the start-up of the experiments. First, the granular parti-
cles (3 mm glass spheres) are filled in the top hopper. The rotating table is accelerated until it
reaches the desired steady rotation rate. Then an experiment is started by opening the outlet
of the hopper by remote control, and the granular material flows in downward direction along
the length of the chute. At the end of the chute, the granular material is collected into the
collection tank. Measurements are made continuously until the hopper is depleted. The time
range during which steady-state flow conditions apply is determined after the experiment by
analysing the time dependence of mass, height and velocity measurements.
Experimental settings and materials
Experiments are performed for angles of inclination of the chute equal to 30 degrees, 40
degrees and 50 degrees with respect to the horizontal. For each angle of inclination we
study four different rotation rates, equal to 0, 4, 8, and 16 rotations per minute (rpm). We
expect the Rossby number to cross unity within this range of rotation rates. The granular
material consists of 3 mm diameter spherical glass particles, purchased from Sigmund and
Linder. These particles are almost monodisperse, with a tolerance of +/- 40 μm. The total
mass of the particles in the hopper is 30 kg. We designed the hopper mouthpiece such that
= 0 rpm = 4 rpm = 8 rpm = 16 rpm
Figure 2.3: Snapshots of granular material flowing down an inclined chute (in the direction
of arrow ) at an inclination angle of 30 degrees for different rotation rates.
a mass flow rate of approximately 1.6 kg/s is achieved, resulting in experimental runtimes of
approximately 20 seconds. The experimental settings are summarized in Table 2.1.
2.3 Results and discussions

Snapshots
It is instructive to consider snapshots of the granular flow through the chute rotating at 0, 4, 8
and 16 rpm shown in Figure 2.3. A clear sideways motion of the particles is observed at the
higher rotation rates. We will study the details of this modified flow in the next subsections.
As we will show later on, maximum downstream velocities at the end of the chute turn out
to be 2.4, 2.9 and 3.25 m/s with the inclination of 30, 40 and 50 degrees and different rota-
tion rates used in the experiments. Table shows the approximate values for the Rossby and
Froude numbers. The Rossby number crosses unity for the lowest angles of inclination and
highest rotation rate studied here. Indeed strong Coriolis effects can be inferred from the
corresponding snapshots. The Froude number is always less than unity, indicating that in all
18 Chapter 2
1.8
1.6
1.4
Mass flow rate (kg/s)
1.2
0.8
0.6
0.4
0.2
0
0 2 4 6 8 10 12 14 16 18 20
Rotation rate (rpm)
Figure 2.4: Effect of rotation on the steady state mass flow rate through the chute at a fixed
angle of inclination of 30 degrees.
cases gravity is dominating the centrifugal forces and able to hold the particles down inside
the chute. In industrial blast furnace, the approximate range of Rossby number is 1 < Ro < 5
and Froude number is 0.05 < Fr < 0.22 .
Mass flow rate

For our study it is important to have a well-controlled steady-state. We have measured the
flow rate at different rotation rates at the exit of the chute by collecting the particles in the
collection tank placed on the dynamic weighing scale. The mass flow rate as a function of
time, as well as the time range of the steady state can thus be determined. Figure 2.4 shows
that there is no significant change in the steady state mass flow rate as a function of rotation
rate, confirming that by placing the hopper at the rotation axis we prevent (considerable)
internal rotational flow inside the hopper which may lead to modified flow rates.
Particle bed height

It is difficult to accurately determine the particle bed height of the flowing granular material
in the rotating chute because 1) the surface of the particle bed is not sharply defined, and 2)
the bed height depends on both width-wise and length-wise position. In the case of a non-
rotating chute, the bed height at the end of the chute is typically a few particle diameters. We
use an ultrasonic sensor of 28 mm in diameter to determine the height of the particle bed,
averaged over a circle of approximately the same diameter. The bed height is measured at
three different width positions, with the sensor centered at 1.5 cm, 4 cm and 6.5 cm from the
side wall (note that the chute width is 8 cm). Height measurements have been performed at 6
0 rpm 4 rpm 8 rpm 16 rpm
x=0.015 m
0.02 0.02 0.02 0.02
φ = 30 deg
x=0.04 m
x=0.065 m
0.01 0.01 0.01 0.01
0 0 0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Bed height(m)
0.02 0.02 0.02 0.02

φ = 40 deg
0.01 0.01 0.01 0.01
0 0 0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
0.02 0.02 0.02 0.02

φ = 50 deg
0.01 0.01 0.01 0.01
0 0 0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length (m)
Figure 2.5: Particle bed height at different positions along the length of the chute for three
different spanwise positions, three different angles of inclination of 30, 40 and 50 degrees,
and four different rotation rates of 0 rpm (non-rotating), 4 rpm, 8 rpm, 16 rpm. Note that the
bed height is averaged (by the height sensor) over a circular disc of diameter 28 mm.
Figure 2.6: Surface plot of the particle bed height in the chute at a fixed angle of inclination
of 30 degrees and rotation rate of (a) 0 rpm (non-rotating) (b) 4 rpm (c) 8 rpm (d) 16 rpm.
20 Chapter 2
length-wise positions, at 0, 10, 20, 30, 50 and 80 cm from the measurement origin, where we
define the measurement origin at 3.5 cm distance (along the chute) from the axis of rotation.
Figure 2.5 shows the bed height as a function of the length-wise position in the chute, for three
different width-wise positions and different angle of inclinations. Without rotation, the bed
height is monotonically decreasing along the length of the chute. Almost the same average
bed height is measured at all three spanwise positions. A similar profile of bed height along
the length of the chute was found literature (Vreman et al., 2007; Holyoake, 2011; Sheng,
2012). As the rotation rate of the chute increases, the bed height increases on the right side
and decreases on the left side of the chute due to the Coriolis forces, which push the particles
towards the right sidewall (looking downstream). At higher rotation rates the bed height even
tends to zero close to the left sidewall at the end of the chute, and a maximum height results at
the right end of the chute, as shown in figure 2.6(d). Experiments at all angles of inclination
( =30, 40 and 50 degrees) show qualitatively similar behaviour. At higher angles, the parti-
cles are accelerated more strongly (as shown in the next subsection), leading to more dilute
flows. Actually, visual inspection learns that as the rotation rate increases at higher angle of
inclination, the bed height increases at the right side of the chute. However, because of the
averaging effect of the sensor area, this is not observable in our height sensor measurements.
More detailed measurements are planned with 3D Particle Tracking Velocimetry to determine
the surface flow and height with more detail.
Figure 2.6 shows a surface plot for the bed height in the chute inclined at 30 deg for different
rotation rates. This plot clearly shows that the bed height changes drastically with increasing
rotation rate.
Surface particle velocity

Figure 2.7 shows the surface velocity field measured at different rotation rates of the chute
inclined at 30 degrees. The color indicates the velocity magnitude. Time averaging has been
done over 1500 pairs of images, corresponding to 15 seconds of steady flow. We observe
that at low rotation rates the velocity vectors are parallel to the length of the chute. However,
at higher rotation rates the velocity vectors obtain a spanwise component too. In the next
subsections we will quantify the average velocities along the length and width of the chute in
more detail.
Streamwise surface velocity along the length of the chute

Figure 2.8 shows the cross-sectional averaged streamwise surface velocity for rotation rates
of 0 and 16 rpm. In making this average, we have excluded (masked) the area which is de-
pleted of particles. We observe that, as the rotation rate increases, there is a slight decrease
in average surface velocity in the upper section of the chute. This may be understood as a
= 0 rpm = 4 rpm = 8 rpm = 16 rpm

Figure 2.7: Velocity vector plots generated by PIV software for surface velocity of particles
on chute inclined at 30 degree and rotation rate of 0 rpm, 4 rpm, 8 rpm and 16 rpm.
consequence of the Coriolis force which drives the particles towards the sidewall, leading to
compaction and an increased friction with the sidewall. In the lower section of the chute,
the average streamwise velocity increases compared to the non-rotating case. Here, the cen-
trifugal forces play an important role, increasingly accelerating the particles with increasing
distance from the axis of rotation. Although these effects are subtle, qualitatively similar re-
sults for 30, 40 and 50 degrees are found. The cross-over point shifts downwards the chute
length as the inclination angle is increased. Streamwise particles velocity and bed height
along the length of the inclined chute (without contraction case) shows similar profiles as
shown in the paper of Vreman et al. (2007). We found similar profiles for granular flows
down non-rotating inclined chutes (Holyoake, 2011; Sheng, 2012).
One may be tempted to think that the streamwise velocity is close to that of a single fric-
22 Chapter 2
φ = 30 deg φ = 40 deg φ = 50 deg

4 4 4
Ω = 0 rpm
Ω = 16 rpm
Averaged streamwise surface particle velocity (m/s)
Ω = 0 rpm single particle

3.5 Ω = 16 rpm single particle 3.5 3.5
3 3 3
2.5 2.5 2.5
2 2 2
1.5 1.5 1.5
1 1 1
0.5 0.5 0.5

0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length (m)
Figure 2.8: Cross-sectional averaged streamwise surface particle velocity along the length
of a chute inclined at three different angles of 30, 40 and 50 degrees for rotation rates of 0
rpm (non-rotating), and 16 rpm. Note that the results of 4 and 8 rpm fall in between these
curves. For comparison, symbols show the streamwise velocity of a single frictionless particle
constrained to the chute plane.
tionless particle falling down an inclined plane. We therefore compare our results with the
trajectory of a single frictionless particle constrained to the chute plane, by solving the equa-
tion of motion (Krcek et al., 1977)
dva
ma Ω × va ) − ma (Ω
= ma g − 2ma (Ω Ω × [Ω
Ω × ra ]) + FN , (2.5)
dt
Here ma is the mass of particle and va is the velocity of the particle in the co-rotating frame,
the second and third terms on the right hand side are the Coriolis and centrifugal forces, re-
spectively, and FN is the reaction force from the chute wall (such that the particle stays within
the plane). The initial position of the particle is chosen at the rotation axis and the initial ve-
locity is chosen equal to that observed in the experiment. Figure 2.8 shows the streamwise
velocity of such a particle up to the point where it reaches the sidewall. Without system
rotation, the experimentally observed velocities are lower than that of a single frictionless
particle, which shows the effects of friction and contact forces. With system rotation, the
experimentally observed velocities are first lower, then higher than that of a single friction-
less particle. The latter is presumably caused by lateral contact forces with neighbouring
particles, which suppress deviations of the particle trajectory to the sides.
Ω = 0 rpm Ω = 4 rpm Ω = 8 rpm Ω = 16 rpm

4 4 4 4
z = 0.0 m
3 3 3 3 z = 0.2 m
φ = 30 deg
z = 0.4 m
2 2 2 2 z = 0.6 m
1 1 1 1
Streamwise surface particle velocity (m/s)
0 0 0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
4 4 4 4
3 3 3 3
φ = 40 deg
2 2 2 2
1 1 1 1
0 0 0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
4 4 4 4
φ = 50 deg
3 3 3 3
2 2 2 2
1 1 1 1
0 0 0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
Chute width (m)
Figure 2.9: Streamwise surface velocity at four length-wise positions (z) as a function of
width-wise position in a chute inclined at an angle, for inclination angles of 30, 40 and 50
degrees.
Streamwise surface velocity along the width of the chute

Figure 2.9 shows the streamwise surface velocity at four streamwise positions as a function
of spanwise position for all 3 inclination angles and 4 rotation rates studied. Without rotation,
the surface velocity is slightly higher at the centre of the chute than near the sidewalls. As the
particles move downwards the velocity increases.
Spanwise surface velocity along the width of the chute

In Figure 2.10 show the spanwise surface velocity across the width of the chute at different
streamwise positions in the chute. At low rotation rates, the spanwise surface velocity is
almost zero because the particles move in parallel streamlines along the length of the chute.
As the rotation rate increases, the spanwise velocity is increasingly influenced by Coriolis
forces. The largest spanwise velocities are observed approximately halfway the chute.
2.4 Conclusions
In this chapter, we have presented an experimental investigation of dry monodisperse granular
flows down an inclined chute for different rotation speeds of the chute. The Froude number is
always less than unity, indicating that in all cases gravity is dominating the centrifugal forces
and able to hold the particles down inside the chute. The observed patterns depend on both
angle of inclination and the rotation speed of the chute. The Rossby number can be used to
24 Chapter 2
Ω = 0 rpm Ω = 4 rpm Ω = 8 rpm Ω = 16 rpm

0.1 0.1 0.1 0.1
z = 0.0 m
0 0 0 0 z = 0.2 m
φ = 30 deg
−0.1 −0.1 −0.1 −0.1 z = 0.4 m

z = 0.6 m
−0.2 −0.2 −0.2 −0.2
−0.3 −0.3 −0.3 −0.3
Spanwise surface particle velocity (m/s)
−0.4 −0.4 −0.4 −0.4

0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
0.1 0.1 0.1 0.1

0 0 0 0
φ 40 deg
−0.1 −0.1 −0.1 −0.1

−0.2 −0.2 −0.2 −0.2
−0.3 −0.3 −0.3 −0.3
−0.4 −0.4 −0.4 −0.4
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
0.1 0.1 0.1 0.1

0 0 0 0
φ 50 deg
−0.1 −0.1 −0.1 −0.1

−0.2 −0.2 −0.2 −0.2
−0.3 −0.3 −0.3 −0.3
−0.4 −0.4 −0.4 −0.4
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
Chute width (m)
Figure 2.10: Spanwise surface velocity at four length-wise positions (z) as a function of
width-wise position in a chute inclined at an angle, for inclination angles of 30, 40 and 50
degrees.
determine whether these rotational effects are significant. With increasing rotation rate the
particles are moving increasingly fast to the right sidewall. The streamwise average surface
particle velocity is slightly reduced in the upper section of the chute due to compaction as a
result of the Coriolis force, but strongly increases in the lower section due to the increasing
importance of the centrifugal forces. The particle bed height becomes a two-dimensional
function of the position inside the chute, with a strong increase in bed height along the right
sidewall due to the Coriolis forces. Both the bed height and width-wise velocity profile show
significant variations in the width direction due to the Coriolis forces. Quantitative data of this
type is still largely lacking in the literature. These measurements provide a well-defined set of
observations suitable for explanation of the main phenomena of the rotating granular flows.
Additionally, they are suitable for confirmation of the Coriolis forces on particle flow and
for refining and validating computer simulations of rotating granular flows. A well-validated
model can give more detailed insight in the flow in the interior region where our experiments
cannot reach. In the next chapter we will show that DEM simulations show quantitative
agreement with experimental measurements of bed height and streamwise surface particle
velocity along the length of chute, both with and without rotation.
Chapter
3
Numerical investigation of
monodisperse granular flow through
an inclined rotating chute ∗
We validate a Discrete Element Model of spherical glass particles flowing down a rotating
chute against high quality experimental data. The simulations are performed in a co-rotating
frame of reference, taking into account Coriolis and centrifugal forces. In view of future ex-
tensions aimed at segregation studies of polydisperse granular flows, several validation steps
are required. In particular, the influence of the interstitial gas, a sensitivity study of the col-
lision parameters, and the effect of system rotation on particle flow is investigated. Chapter
2 has provided the benchmark laboratory measurements of bed height and surface velocities
of monodisperse granular flow down an inclined rotating chute. With a proper choice of the
friction coefficients the simulations show very good agreement with our experimental results.
The effect of interstitial gas on the flow behavior is found to be relatively small for 3 mm
granular particles.
3.1 Introduction
In blast furnaces operated in the metallurgical industry, granular materials such as coke and
pellets are charged from the top of a hopper into the blast furnace through an inclined rotating
chute. Such particulate flows can segregate during the hopper discharge process and on the
rotating chute as a result of differences in material density, size and shape. Thus, a better
∗ This chapter is based on: Shirsath, S.S., Padding, J.T., T.W.J. Peeters, Clercx, H.J.H. and Kuipers, J.A.M.
(2014). Numerical investigation of monodisperse granular flow through an inclined rotating chute. AIChE, Journal,
60(10), 3424-3441.
25
26 Chapter 3
understanding of the (fundamental) phenomena, allowing for a better control of the flow be-
havior of granular materials on the rotating chute, is very important for efficient operation of
a blast furnace. The first step towards generating this understanding is to study monodisperse
granular spherical particles on an inclined rotating chute by employing laboratory experi-
ments and numerical studies of model systems.
Many previous studies have focused on granular flows through non-rotating chutes (Mio et al.,
2009; Zaimi et al., 2009; Yu, 2008), including measurement of the velocity of different types
of granular particles flowing slowly through an inclined open chute (Ishida and Shirai, 1979).
Augenstein and Hogg (1974) investigated the behavior of thin layers of granular particles on
inclined surfaces and measured velocity profiles for different angles of inclination, particle
size and chute surface. They described the observed velocity profiles using a simple frictional
force balance, i.e. their model did not explicitly include bouncing and rolling of the particles.
Therefore, the friction coefficient obtained in the work of Augenstein and Hogg can only be
viewed as an effective one, lumping all effects of bouncing and rolling into a single number. A
disadvantage of such an approach is that the effective friction must be tuned for each particle
size and only applies to shallow granular flows.
To achieve a better understanding of the behaviour of granular flows in rotating chutes, we

have carried out laboratory experiments (bed height measurements with sensors, particle
velocity measurements at the surface of the granular flow with Particle Image Velocime-
try (PIV)) and applied a Discrete Element Model (DEM) for numerical studies. In such a
numerical model each individual particle and its collisions with neighbouring particles is
tracked deterministically in time. The experimental measurements (and applied techniques)
and numerical simulations are strongly complementary as numerical studies give access to
regions of the granular flow hidden from standard optical diagnostics. Before we can exploit
this complimentarity, a strong cross-validation of both methods is required, which will be
addressed in this chapter.
Once validated, DEM models are one of the most useful and reliable simulation tools for the
numerical analysis, flow visualization and in-depth studies of closure relations for granular
flow behaviour during transport (Cundall and Strack, 1979; Weinhart et al., 2012; Thornton
et al., 2012a). DEM can also provide useful information about the optimization of particulate
processes. Several studies on the iron making process have already been published (Nouchu
et al., 2005; Zhou et al., 2005; Mio et al., 2007; Li et al., 2013). An in-house version of our
Discrete Element Method that is usually used for gas-solid fluidized beds (Hoomans et al.,
1996; Van der Hoef et al., 2006; Zeilstra, 2007; Godlieb, 2010; Kou et al., 2013), has been
modified to allow for the simulation of particle flows through a chute rotating around an axis
parallel to gravity.
Numerical investigation of granular flow down a rotating chute 27
DEM simulations have also been extensively applied to study different granular flows in in-
dustries, for example, in drum mixers (Stewart et al., 2001; Kano et al., 2005), for fluidized
beds (Di Maio et al., 2009; Kaneko et al., 1999), for hopper charging and discharging flows
(Nguyen et al., 2009; Yu and Saxén, 2010, 2011a). In the work related to flow in chutes,
Mio experimentally investigated size segregation of sintered ore particles flowing along an
inclined chute for validation of DEM results (Mio et al., 2008). Zhang studied the size segre-
gation of sintering material flows down a rough chute, simulating four types of representative
particles of different diameters and magnetism using DEM modelling (Zhang et al., 2004).
Mio also studied particle behaviour in an inclined rotating chute using DEM and observed
enhanced effective particle velocities in the chute (while moving toward the outlet) due to
the centrifugal force (Mio et al., 2009). Sawley studied how different ways of charging the
granular material onto the vessel leads to different segregation flow and distribution of grain
size using DEM simulations (Sawley et al., 2011). Some recent experimental studies and
DEM simulations of granular flows in inclined rotating chutes and in the hopper was reported
in the papers and thesis of Yu and Saxén (2011b); Yu and Saxen (2012) and Yu (2013).
As mentioned above, the motivation of the current analysis is to carry out a cross-validation
between laboratory measurements and DEM simulations of the granular flow patterns in an
inclined rotating chute. We will perform this cross-validation for two fixed angles of inclina-
tion and different rotation rates of the chute. We study flow of 3 mm spherical glass particles
down an inclined rotating chute by detailed DEM simulations with emphasis on the influence
of the interstitial gas, a sensitivity study of the collision parameters such as restitution and
(rolling) friction coefficients, and the effect of rotation on inlet particle flow. The results are
validated by comparison with data sets from well-defined experiments for a fixed angle of
inclination and different rotation rates of the chute.
This chapter is organised as follows. In section 2 we present a summary of the simulation
model and the parameter settings and in section 3 we briefly explain the computational ana-
logue for calculation of the bed height and surface velocity. Cross-validation with regard to
particle charging, suitable restitution and (rolling) friction coefficients and the influence of
the interstitial gas is discussed in section 4. A direct comparison of DEM simulations with
the experimental data from chapter 2 is presented and discussed in section 5. In section 6
we focus on computational measurements of quantities which are not readily accessible in
optical experiments. Finally we give our conclusions in section 7.
3.2 Simulation model

The discrete element method (DEM) originally developed by Cundall and Strack has been
used successfully to simulate the many granular flow systems (Cundall and Strack, 1979).
28 Chapter 3
Particle collisions are modeled by a soft-sphere contact model, which accounts for the energy
dissipation due to inelastic particle interactions. This energy dissipation is characterized by
the empirical coefficients of normal and tangential restitution and the coefficient of friction.
The particle collision characteristics play an important role in the overall behavior of granular
flows. For this reason, the collision properties of the particles used in our experimental study
were accurately determined by detailed single-particle impact experiments for the normal
and tangential restitution coefficients. We have estimated the coefficient of friction between
particle-wall and particle-particle to achieve good agreement with result for granular flow
in the non-rotating chute at an inclination angle of 30 degrees. Then we validate the model
for other inclination angles and rotating chutes. We will show that with this coefficient of
friction good agreement is found with non-rotating and rotating granular flow experiments.
We will also describe how we have adapted the equations of motion to take into account the
pseudo-forces that appear in a co-rotating frame of reference. For some simulations, we have
included the effects of interstitial gas by coupling the particles to a continuum gas phase.
We will describe the gas phase equations. These equations have been incorporated in our
in-house simulation codes. Let us now start with defining the equations of motion for the
particles.
Equations of motion
In DEM, the particle phase is described by Newtonian equations of motion for each particle
in the system. The translational and rotational equations of motion for a particle a are given
by
dvva
ma = ma g + Fca + Fap + Fda (3.1)
dt
ωa
dω
Ia = Ta + Tr,a , (3.2)
dt
where ma is the mass of particle a, v a is the particle velocity, ω is the rotational velocity and
Ia the moment of inertia around its center-of-mass. The first two terms on the right-hand side
of 3.1 represent, respectively, the forces due to gravity (g is the gravitational acceleration),
and forces due to direct contact with other particles and walls. The particles are coupled with
the surrounding gas phase through the third and fourth term which represents, respectively,
the force due to pressure gradients in the gas phase and the drag force with the gas. Ta is the
torque on particle a, which is determined by the contact forces, and Tr,a is the rolling torque
due to rolling friction, both of which will be explained in the next section.
Figure 3.1: Schematic representation of the linear spring/dashpot soft-sphere model. The
contact force between a pair of particles is decomposed into parallel and tangential com-
ponents. For each component, we define a spring and dashpot to calculate the force. The
transition between a sticking and sliding collision is controlled by the coefficient of friction,
which poses an upper limit to the tangential force.
Soft-sphere contact model

The particle-particle and particle-wall contact force model accounts for the energy dissipated
during collisions, as characterized by empirical coefficients of normal and tangential restitu-
tion, the coefficient of friction, and coefficient of rolling friction. Apart from the rolling fric-
tion, we employ a standard linear spring/dashpot model, in which separate springs and dash-
pots are defined for both normal and tangential displacements (Cundall and Strack, 1979).
A schematic representation of the linear spring/dashpot soft-sphere model is shown in figure
3.1.
In detail, the total contact force on particle a is given as a sum of pair forces with all neigh-
bouring particles with which particle a is overlapping. The overlap (in the normal direction)
between particle a and a neighbour b is defined asδn = Rb + Ra − |rb − ra |. The contact force
on particle a in the normal direction due to this contact is given by
Fn,ab = −kn δn nab − ηn vn,ab (3.3)
where kn is the normal spring stiffness, nab = (rb − ra ) / |rb − ra | is the unit vector pointing
from a to b, ηn is the normal damping coefficient, and vab,n = (vab · nab ) nab is the normal
component of the relative velocity between particles a and b at the point of contact. This
relative velocity is given by
vab = va − vb + (Ra ωa + Rb ωb ) × nab (3.4)
For the tangential component of the contact force, a similar expression is used,
30 Chapter 3
Ft,ab = −kt δt − ηt vt,ab (3.5)
where kt is the tangential spring stiffness, ηt is the tangential damping coefficient, and vab,t =
vab − vab,n is the tangential component of the relative velocity. The tangential overlap δt is
the time integral of this tangential velocity from the time of initial contact (Kou et al., 2013).
The above tangential force, eq. 3.5, applies to the case of a sticking collision where the two
surfaces in contact stick together when the tangential forces are not too large. If, however,
the following relation is satisfied

Ft,ab > μ Fn,ab (3.6)
where μ is the coefficient of friction, then the two surfaces in contact start to slide relative to
each other. In that case the tangential force is limited and replaced by

Ft,ab = −μ Fn,ab tab (3.7)
where tab = vab / |vab |is the unit vector in the tangential direction.
Although it may appear that up to this point we have to define five contact parameters
(kn , kt , ηn , ηt and μ), the number of free parameters is reduced by one by the consistency
requirement that the collision time in the normal direction is the same as in the tangential di-
rection. By using analytic solutions of the damped harmonic oscillator equations for the nor-
mal and tangential directions, it is possible to eliminate, e.g., the tangential spring stiffness kt .
Furthermore, the analytical solutions allow us to calculate the coefficients of restitution for
the normal and tangential directions, defined as en = −vn /vn and et = vt /vt , where the prime
indicates postcollisional relative velocity. In practice we therefore invert the equations, and
use en and et as input parameters, rather than the damping coefficients because the latter are
more difficult to obtain experimentally. Using these parameters, the normal and tangential
collision times are equal to:

mab π 2 + ln2 en
tc = , (3.8)
kn
where mab = ma mb /(ma + mb ) is the reduced mass of the particle pair, and the tangential
spring stiffness is equal to
2 π 2 + ln2 et
kt = kn . (3.9)
7 π 2 + ln2 en
For full details, the reader is referred to Van der Hoef et al. (2006).
The resulting force and torque for particle a are obtained by adding the pairwise contributions
of all the particles b that are in contact with particle a,

Fca = ∑ Fn,ab + Ft,ab (3.10)
b
and

Ta = ∑ Ra nab × Ft,ab . (3.11)
b
We note that contact forces and torques between particles and walls are treated similarly to
particle-particle contacts, but the walls are assumed to be non-moving (in the co-rotating
frame of reference) and of infinite mass.
Sakaguchi et al. first introduced the rolling friction concept into DEM when they performed
a comparison study of experiments and modelling of plugging of granular material during
silo discharge (Sakaguchi et al., 1993). Their rolling friction model was based on an experi-
mental and theoretical study of Beer and Johnson (1973) and proposed by Zhou et al. (1999).
We include a rolling friction in our DEM simulation method to account for the fact that real
particles are never perfectly spherical, or remain spherical upon contact. Indeed, particles in
industrial processes such as in the steel industry tend to have a very irregular shape. The most
accurate solution would be to model the exact particle shape. However, this is computation-
ally very expensive because of the large increase in the number of degrees of freedom, and
more expensive statistical ensemble averaging. Simulating the exact shape is therefore not
suitable for simulating large scale granular flow in industrial processes. The particle shape
mainly affects the particle rotation velocity and the particle packing fraction. As a first ap-
proximation we will still treat the particles as spherical, but with an additional rolling friction
to account for the decrease in rotation velocity (Zhou et al., 1999). To this end we introduce
an additional rolling resistance torque on a particle a due to interactions with a particle b, as
follows
ω rel
Tr,ab = −μr Rr Fn,ab . (3.12)
ω rel |
|ω
ω rel = ω a − ω b (3.13)
where ωrel is the relative angular velocity between particles a and b. The rolling resistance
torque is assumed to scale linearly with the magnitude of the normal force between particles
a and b, and with the rolling radius Rr defined as
Ra Rb
Rr = (3.14)
(Ra + Rb )
The dimensionless coefficient of friction μ from the soft-sphere model and the coefficient of
rolling friction μr from the above model together determine the rolling behaviour of a single
32 Chapter 3
particle falling down a plane inclined at an angle φ . For a perfectly smooth and infinitely stiff
sphere we have μr = 0, and the friction μ at the surface of the sphere causes a pure rolling
motion without sliding for shallow inclination angles tan φ < μ, and rolling and sliding for
steeper inclination angles tan φ > μ. For less ideal, but still nearly-spherical particles, we
have 0 < μr < μ, and the particle will repose (neither slide nor roll) for tan φ < μr , roll for
μr < tan φ < μ, and roll and slide for tan φ > μ. Note that in our work, to ensure a continuous
flow without stagnation, we will keep the inclination angle of the chute fixed at an angle for
which tan φ > μ. However, the above classification is valid only for a single particle falling
down an inclined plane; the effect of rolling resistance on the flow of multiple particles is
more difficult to predict and therefore part of this study.
Also note that in this work we neglect the effect of torsional torque induced by relative rota-
tion of the particles around the line connecting the centers of mass. For details, the reader is
referred to (Luding, 2008; Fuchs et al., 2014).
Coriolis and centrifugal forces

The particles flowing through the chute are in relative motion with respect to the chute bound-
ary, which in its turn is rotating with respect to a fixed frame. Although it is possible to
simulate a moving chute boundary, for pure chute flow it is computationally more efficient
to work in a frame of reference that co-rotates with the rotating chute. The first advantage
is that contact detection between particles and the chute inner wall, as well as computation
of the particle-wall contact forces, can be carried out without the need to find the location
of the chute moving boundary during each computational step. The second advantage is that
it is computationally much cheaper to include the effects of interstitial gas in a co-rotating
frame of reference. Note that for a large scientific community usage of a co-rotating frame
of reference is standard and common practice, e.g. in the meteorological and the geophysical
fluid dynamics community (Van Heijst and Clercx, 2009).
When working in a co-rotating frame of reference, pseudo-forces arise due to the non-inertial
motion of the system. Every individual particle a in the rotating system experiences an ad-
ditional Coriolis force and centrifugal force, both of which must be added to the equation of
motion eq. 3.1,
Ω × va ) − ma (Ω
(F)rotating = (F)stationary − 2ma (Ω Ω × [Ω
Ω × ra ]) (3.15)
where Ω is the angular velocity of the chute and ra the position of particle a relative to a
position located on the axis of rotation. The rotation axis and the directions of the Coriolis and
centrifugal accelerations are all shown in Figure 2.1. Similarly, every particle will experience
an additional Coriolis torque, which must be added to eq. 3.2
Ω × ω a ).
(T)rotating = (T)stationary − Ia (Ω (3.16)
Inclusion of a Coriolis torque is necessary because, even when no other torques apply to a
particle, in the co-moving frame of reference the direction of the particle’s angular momen-
tum will appear to reorient. When applying Eq. 3.16 this reorientation is such that the true
angular momentum (i.e. when viewed from an outside inertial system) is exactly conserved.
We emphasize that the physics does not change when going from an inertial to a co-rotating
frame of reference. The contact forces, which depend on relative positions and relative ve-
locities, remain unaffected. When focusing only on flow through a rotating chute, as we do
in this work, the use of pseudo-forces is simpler than including interactions between particles
and an explicitly moving wall. When the non-rotating process equipment before or after the
chute is also modelled, the advantage of using a co-rotating frame disappears, and using an
inertial frame is preferred.
Equations for gas hydrodynamics

In this work we will test the role of interstitial gas for both non-rotating and rotating chute.
The motion of the gas phase is described by the equations of mass and momentum conser-
vation. Given the large number of particles in DEM simulations one usually resorts to the
volume-averaged Navier-Stokes equations. From mass conservation we have
∂ (ερg )
+ ∇ · (ερg u) = 0 (3.17)
∂t
where ρg is the gas density, ε is the local porosity and u is the gas velocity measured in the
co-rotating frame. The momentum equation for the gas phase reads
∂ (ερg u)
∂t
Ω × r)
+ ∇ · (ερg uu) = −2ερg Ω × u − 2ερg Ω × (Ω
(3.18)
−ε∇p − S p − ∇ · (ετg ) + ερg g
where p is the gas pressure, S p is a source term caused by drag with the particles (see below)
and g is the gravitational acceleration. Note that the first two terms on the right hand side
of eq. 3.18 take into account the effect of system rotation on vectorial quantities (in this
case velocities) (Kundu and Cohen, 2008), but that no such terms appear in the continuity
equation eq. 3.17 because mass is a scalar quantity. The viscous stress tensor for the gas
phase is assumed to obey the general form for a Newtonian fluid, i.e.
2
τg = −(λg − μg )(∇ · u)I − μg (∇u + (∇u)T ) (3.19)
3
34 Chapter 3
where λg is the gas phase bulk viscosity, μg is the gas phase shear viscosity, and I the unit
tensor.
To solve the above equations, the chute is divided into NX x NY x NZ computational cells
(in the co-rotating frame of reference). The pressure and gas velocity in each cell can be
calculated from discretized versions of equations 3.17 and 3.18 (Goldschmidt et al., 2001).
The sink term S p in eq. 3.18 represents the drag experienced by the particles in the com-
putational cell, which we treat as point sources of momentum for the gas phase. In detail
(Goldschmidt et al., 2001)
ˆ
1 βVa
∑a=1
N part
Sp = (u − va )δ (r − ra )dV (3.20)
V 1−ε
where V represents the local volume of the computational cell and Va is the volume of particle
a. The distribution function δ distributes the reaction force acting on the gas phase to the
computational cells, where a trilinear interpolation method is used in this work (Hoomans
et al., 1996). The inter-phase momentum transfer coefficient, β , describes the drag of the gas-
phase acting on the particles, which is modelled using the correlation proposed by Beetstra
et al. (2007):
(1 − ε)2 μg
β =A + B(1 − ε)Re
ε da2
18ε 4 √
A = 180 + 1 + 1.5 1 − ε (3.21)
1−ε

0.31 ε −1 + 3ε (1 − ε) + 8.4{Re−0.343
B=
1 + 103(1−ε) Re2ε−2.5
where Re = ερg |u − va | da /μg is the particle Reynolds number and is da = 2Ra the particle
diameter.
The pressure and drag forces acting on the particles in eq. 3.1 are given by
Fap = −Va ∇p (3.22)
Va β
Fda = (u − va ) (3.23)
1−ε
Note that total momentum of the gas and solid phase is locally conserved: eq. 3.22 is con-
sistent with the third term on the r.h.s. of eq. 3.18, and eq. 3.23 balances the source term in
eq. 3.20.
The boundary conditions for the gas phase imposed for the side walls and the bottom wall
of the chute is no-slip and for the top boundary we impose a prescribed velocity in all three
directions equal to u = −ΩΩ × (r − raxis ), representing the position-dependent cross-wind (lid-
driven flow) present at the open chute top during the experiments. We emphasize that the lat-
ter boundary condition represents a worst-case scenario for the influence of the gas because
(1) with respect to gas flow in the widthwise direction we assume a maximum amount of lid-
driven flow, while in reality the air near the open top boundary of the chute will not remain
stagnant in the inertial lab-frame, and (2) because with respect to the gas flow in the stream-
wise direction we assume a maximum amount of deceleration caused by the top boundary
(near-zero z-velocity) instead of a stress-free condition (with zero velocity gradient). A co-
flow velocity inlet boundary condition is imposed at the inlet and a constant pressure outlet
boundary condition at the exit of the chute.
Simulation settings
The physical properties of the spherical glass particles and the conditions for the simulation
are shown in Table 3.1. Note that the normal spring stiffness is chosen lower than the real
value for a 3 mm glass particle, because using a realistic value would lead to a prohibitively
small integration time step in the simulation. It is common practice in DEM simulations to
use a lower value of the spring stiffness, which is allowed as long as the normal overlap is
kept small (maximum 2% of the particle diameter) and the collision time is still small relative
to times on which particle configurations change.
The flow domain was divided into small cells, with the size of each cell approximately twice
the particle diameter. This cell size has been shown to yield accurate results for the particle-
gas interaction forces in simulations of dense fluidized beds (Laverman, 2010).
In our DEM simulations the chute is initially empty, as is the case in the physical experiments.
Simulations are carried out for a constant mass flow rate at the inlet of the chute. We introduce
the particles at the chute entrance in a rectangular area with a variable height, which we refer
to as the sluice height. We arrange the particles in a bcc-lattice using as many particles as
can be fitted in the sluice height. Because the mass rate is fixed, the streamwise velocity with
which the new particles are initialized is smaller when the sluice height is larger. Specifically,
we introduce the particles with a mass flow rate of 1.6 kg/s, which is equal to the mass flow
rate of our previous experiments (chapter 2). The total number of particles in the chute
depends on the inclination angle and rotation rate and also fluctuates in time. The typical
number was approximately 31.000 for an inclination angle of 30 degree and approximately
26.000 for 40 degree. The simulations were carried out on a single core of an Intel Xeon
E5520 processor (at 2.27 GHz). In general each simulation required 60 hours of calculation
time for each 6 seconds of simulation.
36 Chapter 3
Table 3.1: Simulation setting for monodisperse flows of glass particles).
Property Value
Number of cell x-direction 16
Number of cell y-direction 15
Number of cell z-direction 150
Total number of particles 31,000
Particle type Spherical glass
Particle density 2550 kg/m3
Normal spring stiffness 1000 N/m
Coefficient of normal restitution en,pp = en,pw = 0.96
Coefficient of tangential restitution et,pp = et,pw = 0.33
Coefficient of friction μ pp = 0.10, μ pw = 0.22
Coefficient of friction μr,pp = 0.0, μr,pw = 0.0
Total simulation time 6.0 s
Time step 2.5 × 10−6 s
3.3 Computational measurements

Calculation of bed height in simulations
We calculated the bed height in our simulations similarly to the bed height measurments in the
experiments described in chapter 2. In the experiments, an electronic height sensor was used
which measures the height by averaging the top of the highest particles over a circular area
with a diameter of 28 mm. Independent measurements on stationary packed beds showed that
the experimental error in these bed height measurements is about 50% of a particle diameter,
i.e. 1.5 mm. We measured the averaged bed height at three different width-wise positions,
i.e. the left, center and right side of the chute, centered at 0.015m, 0.045m and 0.065 m of
the 0.08 m wide chute. In our simulation, the time averaged bed height was calculated over a
matrix of 5 x 5 cells with similar dimensions as the sensor surface, as shown in figure 3.2.
Calculation of surface velocity in simulations

In the experiments described in Chapter 2, we used Particle Image Velocimetry (PIV) to de-
termine the average two-dimensional velocity field of the visible surface particles (Keane and
Figure 3.2: Top view of a section of the chute with the grid cell structure used for bed height
calculation. Chute walls are present at the left and right side of this figure and the granular
flow is in the Z direction. In the experiments, the bed height is averaged over a circular area
at three width-wise positions: left, center and right (dashed circles). For validation, in our
simulations the bed height is averaged over approximately the same areas consisting of 5x5
cells, indicated by red, green and yellow color, respectively.
Adrian, 1991; Westerweel, 1997). In this chapter, we will validate our DEM simulations by
comparing particle velocity measurements between experiments and simulations. To enable
a direct comparison, the two-dimensional particle velocity field is calculated for those parti-
cles that are optically visible from above the chute, because in the experiments the camera
was mounted perpendicularly above the chute at a large distance of 1.5m. In our simulations,
the surface velocity is calculated by time-averaging the average velocity of the topmost 6
particles in each streamwise and with spanwise cell. This choice closely matches the number
of experimentally visible particles because at the given particle diameter of 3 mm, and the
width-wise and length-wise size of the cells of 5 mm x 6 mm, it is expected that 4 to 6 par-
ticles will be visible from above the chute. We have checked the dependence of the surface
velocity measurements on the number of topmost particles used, and found only a negligi-
ble influence (less than 0.1% difference in measured surface particle velocity) for particle
numbers ranging from 3 to 12.
3.4 Results and discussion

We will now compare results obtained from the DEM simulations with the experimental
results of chapter 2. Specifically, we will focus on the particle bed height and streamwise and
spanwise particle velocities at the surface of the granular flow for different rotation rates and
fixed angles of inclination of 30 and 40 degrees. First the effect of charging at the inlet of the
chute, and subsequently the influence of the interstitial gas on the granular flow behaviour
is studied. We then report on a sensitivity study to check the effect of the particle collision
parameters on particle bed height in the chute.
38 Chapter 3
0.05
sluice height = 0.025 m
sluice height = 0.035 m
0.045 sluice height = 0.045 m
0.04
0.035
Bed height (m)
0.03
0.025
0.02
0.015
0.01
0.005
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Chute length(m)
Figure 3.3: Time-averaged bed height along the length of the chute inclined at 30 deg for dif-
ferent sluice heights feeding a non-rotating chute. The particle mass loading rate is constant
and equal to 1.6 kg/s. In all subsequent figures a sluice height of 0.045 m is used, and the
origin of the horizontal axis is shifted to obtain a match between experiment and simulation
in a single point, as described in the main text.
Influence of inlet particle charging

In real-life blast furnace processing, the particles are charged from a hopper onto the sur-
face of the chute. In our simulations we can pour the particles in a similar way. However,
we expect that the granular flow through the chute is primarily determined by the particle
properties and the particle mass loading rate. In other words, we expect that memory of the
details with which the particles are introduced into the chute will fade quickly relative to the
time needed to flow down the chute. To confirm this, we introduce the particles at the inlet of
the chute in a rectangular area with a variable height which we refer to as the sluice height.
We arrange the particles in a bcc-lattice using as many particles as can be fitted in the sluice
height. Because the mass rate is fixed (to 1.6 kg/s), the streamwise velocity with which the
new particles are initialized is smaller when the sluice height is larger. We performed several
sets of simulations for different sluice heights feeding a non-rotating chute.
Figure 3.3 shows that the bed height is indeed independent of the sluice height (and therefore
independent of initial velocity) after approximately 0.2 m from the inlet of the chute, espe-
cially when the sluice height is more than 3 cm. The same holds true for the particle velocity.
We have also tested the influence of sluice height for an inclination angle of 40 degrees and
for rotating chutes, rotating at the maximum rate of 16 rpm. Moreover, we performed a series
of simulations where the particles are introduced in the same manner as in the experiments,
i.e. from a hopper at the top directly onto the chute surface (which are computationally more
expensive simulations). For all these cases we found that, again, the bed height and parti-
cle velocity are independent of sluice height after approximately 0.2 m from the inlet (not
shown). In all subsequent simulations we have used a constant sluice height of 0.045 m.
Conversely, the bed height and surface flow velocity do depend on the manner of particle
introduction in the first 0.2 m of the simulated chute. To be able to compare experiments
and simulations, we have determined the first point at which the bed height and surface flow
velocity match between experiment and simulation for a single non-rotating chute at an angle
of inclination of 30 degrees. In all subsequent figures, we have shifted the origin (for distance
along the flow direction) to this point. It is important to have exactly the same distance to the
rotation axis in both simulations and experiments. In detail, in both our simulations and in
our experiments the axis of rotation is cutting through the bottom plane at a distance of 0.035
m before the point of origin determined above.
Influence of interstitial gas
A granular material flow is defined as the flow of a fluid–particle system in which the ef-
fects of the interstitial fluid are assumed to be negligible. This is the asymptotic state of
particle-fluid flows in which the majority of the momentum is transferred by particle-particle
interactions. Intuitively this implies that: (1) the fluid viscosity is low, minimizing the effect
of viscous drag on the particles, and (2) that the solid density is much higher than the fluid
density, so minimizing both relative buoyancy (gas-solid) and added mass effects. To study
the influence of the interstitial gas for our system, we performed simulations with and without
gas. In the simulations with gas both the particles and the gas are introduced at a constant
flow rate. Figure 3.5 shows snapshots of monodisperse flows down chute with and without
gas for rotating case (16 rpm).
Figure 3.4 shows the average particle velocity in both a non-rotating chute (0 rpm) and a
rotating chute (16 rpm), which increases from a low value at the inlet to a large value at the
outlet as a result of gravity. When gas is included in the simulation, the averaged velocity
decreases slightly in both the non-rotating and the rotating chute. However, the influence of
the interstitial air is relatively minor, lowering the velocity at the end of the chute by 2-3% in
a nonrotating chute and 5-10 % in a rotating chute (at both angles of inclination studied in this
work). We will show in the remaining part of this work that this small effect can effectively
and efficiently be incorporated by tuning the coefficient of friction, leading to an accurate
agreement with experimentally observed particle velocities and bed height.
40 Chapter 3
Ω = 0 rpm Ω = 16 rpm
2.5 2.5
Streamwise averaged particle velocity (m/s)
2 2
1.5 1.5
without drag without drag

with drag with drag
1 1
0.5 0.5
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length (m)
Figure 3.4: Time-averaged streamwise particle velocity in a chute inclined at 30 degrees,

comparing simulations with and without air drag.
Sensitivity study of collision parameters

To investigate the sensitivity of the flow on the collisional properties of the particles, we
have performed a series of DEM simulations in which the collisional parameters were sys-
tematically varied from the base case (for glass particles) given in Table 3.1. Each time one
particle-particle collision parameter was varied and its influence on the surface velocity of
the particles was determined.
Base case simulation: glass particles

The base case simulation was performed for the flow of monodisperse glass spheres down a
chute inclined at 30 degree. We note that at first we used in our simulations the coefficient of
friction between glass particles and between glass particles and plexiglass walls as measured
by single particle experiments in our lab. We found that the coefficient of friction between
glass particles is 0.1189 and between glass particle and plexiglass wall is 0.097. These values
are close to the values reported in the literature (Alizadeh et al., 2014). Similar to findings
in the thesis of Goldschmidt (2001). We found that we actually needed to use a higher value
for the friction coefficient between particle and wall (0.22 instead of 0.097) to obtain quan-
titative agreement with the experimental (non-rotating) chute flow results. This higher value
is compensating for the fact that smaller values of the materials’ Young’s modulus (lower ef-
fective spring stiffnesses) are used in the simulations to allow for reasonable time integration
Figure 3.5: Snapshots of steady-state granular flows, flowing from top to bottom through a
chute inclined at 30 degrees and rotating at 16 rpm. Particle are color-coded by their velocity.
Top: flow without gas drag. Bottom: flow with gas drag included in the simulation; arrows
show the gas velocity vectors above the particle bed.
42 Chapter 3
3
Averaged surface particle velocity (m/s)
2.5
1.5
Experiment @ Ω = 0 rpm
Base case simulation @ Ω = 0 rpm
0.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length(m)
Figure 3.6: Time-averaged streamwise surface particle velocity along the length of the chute
inclined at 30 deg. Base case simulation for the collision properties of glass spheres. Small
circles represent experimental data and the line is from the base case simulation.
steps. To be precise: in our work, the value for the particle-wall friction coefficient has been
tuned once to achieve maximum agreement between the simulated and experimental surface
velocity and bed height for the non-rotating base case at 30 degrees chute angle. All other
cases are then used as validation of our model. The surface velocity of the particles obtained
from the base-case simulation is compared with the experimental surface velocity in Figure
3.6. We observe that the simulated time-averaged surface velocity is in good agreement with
the experimental surface velocity at the same operating conditions.
Effect of tangential restitution coefficient
We first study the influence of the coefficient of tangential restitution (for particle-particle
and particle-wall collisions) on the averaged surface velocity in the chute. We varied the
tangential restitution coefficient et from 0.1 to 0.9, where the base case value is 0.33, while
keeping et,pw equal to et . In all cases we obtained results which are nearly indistinguishable
from the base case shown in figure 3.6 (and are therefore not shown). This is already a first
indication that in our systems the tangential forces between the particles are usually saturated
to values limited by the coefficient of friction. We will return to this point when we investigate
the influence of the coefficient of friction.
3
2.5
1.5
μpp =0.12 , μpw=0.22
μpp =0.06 , μpw=0.11
1 μpp =0.24 , μpw=0.44
0.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length(m)
Figure 3.7: Time-averaged streamwise surface particle velocity along the length of chute
inclined at 30 degrees for different coefficients of friction. Small circles are experiments
using glass and lines are simulations (where the green line is the base case for glass).
Effect of normal restitution coefficient

Next we study the influence of the coefficient of normal restitution for both particle-particle
and particle-wall collisions. We varied en from 0.3 to 0.99, where 0.97 is the base case, while
keeping the ratio en,pw /en,pp constant. Under certain conditions the coefficient of normal
restitution is a sensitive parameter for the flow behaviour in DEM studies (Müller et al.,
2008, 2009). However, for chute flow we find only a very weak dependence of the average
velocity on en (not shown). This is in contrast with e.g. the flow of particles in a fluidized
bed, where small perturbations in solid fraction grow exponentially, reinforced by the gas
flow and inelastic collisions. Such a formation of heterogeneous structures does not occur in
the chute flows studied here.
Effect of coefficient of friction

The coefficient of friction is important because it determines the maximum tangential force
that can be exerted at the surface of a particle for a given amount of normal force. As such, it
determines the transition from a sticking to a sliding collision. We performed several simu-
lations using different values of the coefficient of friction, with a fixed ratio for particle-wall
collisions, while keeping all other collision parameters constant. The result presented in Fig-
ure 3.7 show that the streamwise surface particle velocity is very sensitive of the value of the
coefficient of friction. With increasing, the surface velocity decreases continuously. From
this we conclude that the particles are exerting relatively large tangential forces on each other
44 Chapter 3
3
2.5
1.5
μr,pp =0.0 , μr,pw=0.0
μr,pp =0.01 , μr,pw=0.02
1 μr,pp =0.1 , μr,pw=0.2
0.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length(m)
Figure 3.8: Time-averaged streamwise surface particle velocity along the length of chute in-
clined at 30 degrees for different coefficients of rolling friction. Small circles are experiments
on glass and lines are simulations (the green line is the base case for perfect glass spheres).
and are considerably slowed down if their surfaces are rough. The transition between sticking
and sliding collisions, determined by the coefficient of friction, is therefore one of the most
important phenomena in particle chute flow.
Effect of coefficient of rolling friction

Finally, we study the effect of the coefficient of rolling friction r (for both particle-particle
and particle-wall collisions) on the particle surface velocity in the chute. The value of the
rolling friction was varied between zero and its maximum which can be reasonably assumed
for near-spherical particles, i.e. a maximum equal to the coefficient of friction , while keeping
all other parameters constant as shown in Table 3.1. Figure 3.8 shows that the influence of the
coefficient of rolling friction on the surface particle velocity is relatively minor, except when
it is close to the coefficient of friction. With increasing rolling friction, the surface velocity
decreases slightly, except in the initial part of the chute where the particles do not yet rotate
significantly, and are therefore not yet influenced by rolling friction.
Summary of sensitivity study

In summary, for the type of granular chute flow considered here, we have found that the
streamwise surface particle velocity is very sensitive to the value of the coefficient of friction,
somewhat sensitive to the value of the coefficient of rolling friction, and insensitive to the
values of the coefficients of normal and tangential restitution. It is therefore most important
Figure 3.9: Top view snapshots of steady-state granular flows, flowing from left to right
through a chute inclined at 30 degrees. The chute is rotating with a rotation rate of 0 rpm,
8rpm or 16 rpm. Greyscale: experiments. In the simulations snapshots, particles are color-
coded according to the magnitude of their streamwise velocity, from blue to red for low to
high velocity.
to choose the coefficients of friction between particles and between particles and walls cor-
rectly. The right choice can only truly be confirmed by comparing simulation predictions
with experiments for a range of properties and a range of experimental conditions, which is
the topic of the next section.
3.5 Validation: comparison with experimental results

In the previous section we have confirmed that our DEM simulations are able to predict
the steady-state streamwise surface particle velocity of an experimental system consisting of
glass spheres of 3 mm flowing down a non-rotating chute inclined at 30 deg. The reader is
reminded that the parameters for this base case are given in Table 3.1. In the remainder of
46 Chapter 3
= 0 rpm = 4 rpm
0.03 0.03
0.02 0.02
0.01 0.01
Bed height (m)
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
= 8 rpm = 16 rpm
0.03 0.03
0.02 0.02
0.01 0.01
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length (m)
Figure 3.10: Bed height as a function of streamwise position (along the length of the chute)
for three different width-wise positions in a chute inclined at 30 degrees, rotating with rates
of 0 rpm, 4 rpm, 8 rpm and 16 rpm. Small circles are experiments and lines are simulations.
Note that the bed height is averaged over an area equal to the height sensor surface area.
this chapter, we validate the model for more chute settings, with chutes rotating at different
rotation rates and two different inclination angles, comparing not only the streamwise surface
particle velocity, but also the spanwise surface particle velocity and particle bed height profile.
The major effect of chute rotation is a sideways deflection of the particle stream due to Cori-
olis forces present in the frame of reference co-rotating with the chute. This is confirmed
in Figure 3.9, which shows snapshots of the steady-state flow of monodisperse particles in a
chute inclined at 30 degrees for different rotation rates, where colors indicate the magnitude
of the velocity of the particles.
Particle bed height

Figure 3.10 shows the bed height as a function of position along the length of the chute, for a
chute inclined at a fixed angle of 30 degrees. Figure 3.11 shows the same measurements for
an inclination angle of 40 degrees. The bed height is averaged over an area corresponding
to the area of the ultrasonic height sensor. The simulation results (lines) are compared with
experimental results (symbols) for three different width-wise positions in the chute and six
different length-wise positions. For the non-rotating chute the bed height continuously de-
creases along the length of chute, and the results are nearly indistinguishable for the different
width-wise positions. As the rotation rate of the chute increases, the bed height increases on
the right side of the chute and decreases on the left side. Generally, given the experimental
error of approximately 1.5 mm, the simulation results are in good agreement with the ex-
= 0 rpm = 4 rpm
0.03 0.03
0.02 0.02
0.01 0.01
Bed height (m)
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
= 8 rpm = 16 rpm
0.03 0.03
0.02 0.02
0.01 0.01
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length (m)
for three different width-wise positions in a chute inclined at 40 degrees, rotating with rates
of 0 rpm, 4 rpm, 8 rpm and 16 rpm. Small circles are experiments and lines are simulations.
Note that the bed height is averaged over an area equal to the height sensor surface area.
perimental measurements. There is one exception: at higher rotation rates of 8 rpm and 16
rpm the experiments show a maximum in the height as a function of streamwise position at
the right side of the chute (blue symbols), which is much less apparent in our simulations
(blue lines). We attribute this to our previous observation that the precise manner of particle
introduction is important for the first 0.2 m. At the highest rotation rate of 16 rpm memory of
the precise manner of introduction appears to extend somewhat further down the chute, but
agreement between experiments and simulations is found again beyond 0.4 m.
Streamwise surface velocity in the width-wise direction
Figures 3.12 and 3.13 show the streamwise surface particle velocity as a function of width-
wise position at four different streamwise positions, for rotation rates 0 rpm (non-rotating
chute), 4 rpm, 8 rpm and 16 rpm. The streamwise positions are at z = 0.0 m, 0.2 m, 0.4 m
and 0.6 m. The simulation results (lines) are compared with experimental results (symbols).
The figures show that in all cases the streamwise velocity is slightly lower near the side walls
of the chute. Moreover, as the rotation rate increases, the particles move sideways causing
a lack of data at higher width-wise positions. Again we find good agreement between the
simulations and experimental results.
48 Chapter 3
= 0 rpm = 4 rpm
3 3
2 2
1 1
0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
= 8 rpm = 16 rpm
3 3
2 2
1 1
0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
Chute width (m)
Figure 3.12: Streamwise particle velocity along the width of the chute inclined at 30 degrees
for a rotation rate of 0 rpm, 4 rpm, 8 rpm and 16 rpm. Symbols represent the experiments,
lines are results for simulations.
= 0 rpm = 4 rpm
3 3
2 2
1 1
0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
= 8 rpm = 16 rpm
3 3
2 2
1 1
0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
Chute width (m)
Figure 3.13: Streamwise particle velocity along the width of the chute inclined at 40 degrees
for a rotation rate of 0 rpm, 4 rpm, 8 rpm and 16 rpm. Symbols represent the experiments,
lines are results for simulations.
= 0 rpm = 4 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
= 8 rpm = 16 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08

Chute width (m)
Figure 3.14: Spanwise particle velocity along the width of chute for an inclination angle of
30 degrees for a rotation rate of 0 rpm, 4 rpm, 8 rpm and 16 rpm. Symbols represent the
experiments, lines are results from simulations.
= 0 rpm = 4 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
= 8 rpm = 16 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08

Chute width (m)
Figure 3.15: Spanwise particle velocity along the width of chute for an inclination angle of
40 degrees for a rotation rate of 0 rpm, 4 rpm, 8 rpm and 16 rpm. Symbols represent the
experiments, lines are results for simulations.
50 Chapter 3
Spanwise surface velocity in the width-wise direction

Figure 3.14 illustrates the spanwise surface particle velocity along the width of the chute
at different cross sections in the length of chute. The simulation results are compared with
experimental results for different rotation rates of the chute at an inclination of 30 degrees.
We observe that the magnitude of the spanwise velocity first increases and then decreases
for consecutive streamwise positions. This corresponds to the process of sideways motion
induced by Coriolis forces, which is finally stopped by the compaction of the granular flow
against the side wall. The maximum spanwise velocity increases with increasing rotation
rate. Similar results, but with even larger spanwise velocities are found for a chute inclination
angle of 40 degrees, as shown in figure 3.15. Again some deviations between simulation and
experiments are observed at the start of the chute (z = 0 m, blue symbols and blue lines)
at higher rotation rates, consistent with our previous observation that the precise manner of
particle introduction is important for the first 0.2 m. Taking this into account, rather good
agreement is found between experiments and simulations.
3.6 Computational measurements of optically inaccessible properties

We have used the experiments on bed height and surface particle velocity to validate the DEM
simulations. We now give two examples of how the simulations can be used to obtain insight
in properties that cannot readily be obtained from optical experiments.
Depth dependent flow profiles

The first example concerns the depth dependent shear flow profile of a granular flow, i.e. the
streamwise velocity as a function of depth.
In figure 3.16 we show the flow profiles for non-rotating (0 rpm) and rotating (16 rpm) chutes
at different streamwise positions. The results for the non-rotating case have been averaged
over the full width of the chute, whereas for the rotating case we have made two measure-
ments, one averaging over the left half of the chute width and another averaging over the
right half (excluding cells without particles). Figure 3.16 shows that in all cases the granular
material behaves almost like a plug flow, with only a relatively small velocity gradient. In
other words, the chute flow is dominated by slip with the bottom wall. This is consistent with
our previous observation of the dominance of the coefficient of friction.
In the rotating chute, particles in the right half of the chute (as seen in the direction of flow)
are slightly slowed down relative to a non-rotating chute. This may be caused by a dominance
of the additional friction with the side wall induced by Coriolis forces over the centrifugal
forces. Particles in the left half of the chute are not pushed against a side wall, and are
0.03
0.025
0.02
Height (m)
0.015
0.01
0.005
0
0 0.5 1 1.5 2 2.5
Streamwise particle velocity ( m/s)
Figure 3.16: Average streamwise particle velocity along the depth of the chute for an inclina-
tion angle of 30 degrees. Solid lines: non-rotating chute. Dotted lines: left half of the chute
for 16 rpm; dashed lines: right half of the chute for 16 rpm.
therefore influenced more by the centrifugal forces, leading to a slight acceleration relative to
a non-rotating chute.
Depth dependent solid volume fraction

The second example of a property that is difficult to access experimentally is the depth de-
pendent solid volume fraction. Figure 3.17 shows the contour plots of solid volume fraction
as a function of the depth of the chute at different three widthwise positions. The simulation
results shows that the bottom part of the granular flow in a chute is very high, but the solids
volume fraction decreases towards the free surface, with the largest gradient near the free
surface. The solid fraction is slightly different along the center line compared to the side po-
sitions for the non-rotating chute. For the case of rotation the particles moves sideways which
results in an increase in solid volume fraction on the right side of the chute. Measurements
such as these are very difficult to obtain experimentally and show the complimentary strength
of DEM simulations.
3.7 Conclusions
In this chapter, we have presented a comparative study of DEM simulations and experimental
results of dry granular flows of 3 mm glass spheres down an inclined rotating chute. The key
objective of the work was to validate the model for granular flow on a rotating chute with
experimental results of the bed height and surface particle velocity in the chute. We found
52 Chapter 3
= 0 rpm
0.6
0.04 x = 0.015 m
0.03 0.4
0.02
0.5 0.1
0.2
0.3
0 0
0.2
0.01 0.4 0.1
0.2
0.3
0.4
0.1
0.2
0.3
0.6 0.5 0.4
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0.6
0.04 x = 0.045 m
0.03 0.4
0.5
0.02
0.1
0.2
0.3
0 0
0.2
0.01 0.4
0.5
0.1
0.2
0.3
0.4
0.1
0.2
0.3
0.4
0.6 0.5
05 0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0.6
0.04 x = 0.065 m
0.03 0.4
0.02
0.5
0.1
0.2
0.3
0 0
0.2
0.01 0.4 0.1
0.2
0.3
0.4
0.1
0.2
0.3
0.6 0.5 0.4
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
(a)
Height (m)
= 16 rpm
0.6
0.04 x = 0.015 m
0
0.03 0
0.4
0.1 0.1
0.1 0.2
0.02 0.2
0.3
0.4 0.3
0.4
0.2
0.3
0.5
0.5 0.4 0.2
0.01 0.5
0.6
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0.6
0.04 x = 0.045 m
0.03 0.4
0.5
0.2
0.02 0 0
0.4 0.1
0.3
0
0.2
0.01 0.2 0.1
6
0. 0.5 0.4 0.3 0.1
0.2
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0.6
0.04 x = 0.065 m
0.1
0.03 0.4
0.02
0 0.2
0.01 0.4
0.3 0.2
0.5
0.6
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length (m)
(b)
Figure 3.17: Contour plots of average solid volume fraction along the depth of the chute for
an inclination angle of 30 degrees and rotation rate of 0 rpm (a), and 16 rpm (b). Averages
are taken over slices with a width of 0.025m located at the left, center and right side of the
chute.
that the precise inlet conditions are very important for the flow behaviour in the initial part of
the chute, but have no influence on the flow behaviour after a relatively short distance from
the inlet (in our case about 20 cm). We also found that the effect of air on the flow behavior
of the particles is relatively small and can be captured effectively and efficiently by tuning the
friction coefficient. For this type of chute flow, the most important particle contact parameters
are the friction coefficient and, to a lesser extent, the rolling friction coefficient. The precise
values of the normal and tangential coefficients of restitution have no discernible effect on
the particle flow. The main effect of rotation is obvious, namely to push the particle stream
against the side wall. However, in this work we have shown that the simulation model is also
capable of quantitatively predicting the experimental results. We have highlighted how the
simulation model can subsequently be used to predict properties that are not readily accessible
to optical experiments. We conclude that these simulations can be used with confidence to
obtain more detailed insights in complex flows such as mixtures of particles of different size
or different density. This will be the topic of our further study.
Chapter
4
Cross-validation of 3D particle
tracking velocimetry for the study of
granular flows in rotating chutes ∗
Three-dimensional particle tracking velocimetry (3D-PTV) is a promising technique to study
the behaviour of granular flows. The aim of this chapter is to cross-validate 3D-PTV against
independent or more established techniques, such as particle image velocimetry (PIV), elec-
tronic ultrasonic sensor measurements for bed height, and the discrete element model (DEM),
for the complicating circumstance in which granular particles are flowing down a rotating
chute. 3D-PTV can be used to gain access to Lagrangian trajectory data of surface particles
in such flows, from which independent measurements of both the surface velocity and the bed
height in the chute can be derived. The 3D-PTV method is based on imaging and tracking
colored tracer particles that are introduced in the granular material, which are viewed from
three directions. The three cameras collect consecutive frames a known Δt apart and the PTV
algorithm for locating and tracking particles is used to determine particle trajectories and
velocities. We find that the 3D-PTV results are in good agreement with PIV results with regard
to the streamwise and spanwise surface velocity of particles in the rotating chute. The parti-
cle bed height obtained from 3D-PTV is also found to be in good agreement with data from an
ultrasonic bed-height sensor. The experimental findings from PTV for the non-rotating case
are used to further tune the friction coefficient in our discrete element model (DEM) sim-
ulations. The simulation method is validated by the good agreement between experimental
findings and simulations at all rotation rates studied.
∗ This chapter is based on: Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. (2014) Cross-
validation of 3D Particle Tracking Velocimetry with other technique for granular flows down rotating chutes. Sub-
mitted to CES Journal.
55
56 Chapter 4
4.1 Introduction
Experimental studies of granular flows are difficult to perform due to the opaque nature of
such materials. Nevertheless, insight has been generated through use of force sensors (Savage
and McKeown, 1983; Zenit et al., 1997), acoustic probes (Bennett and Best, 1995), tracked
transmitters (Dave et al., 1999), capacitance probes (Louge et al., 1997), optical sensors (Dent
et al., 1998), positron emission particle tracking (Ding et al., 2001; Parker et al., 1997), ra-
dioactive particle tracking (Alizadeh et al., 2013; Sherritt et al., 2003), magnetic resonance
imaging (Kawaguchi, 2010; Hill et al., 1997), digital imaging (Guler et al., 1999; Capart
et al., 2002; Bonamy et al., 2002), fiber optic probes (Boateng and Barr, 1997), particle im-
age velocimetry (PIV) applied to 2D granular flows (Bokkers et al., 2004; Laverman et al.,
2008; Zeilstra et al., 2008; Deng and Wang, 2003) and particle tracking velocimetry (PTV)
(Chou and Lee, 2009; Yang and Hsiau, 2006; Liao and Hsiau, 2009; Jasti and Higgs III,
2008; Veje et al., 1999; Jain et al., 2002). In the past a large amount of experimental work has
been performed on granular flows through inclined channels, see e.g. (Augenstein and Hogg,
1978; Brennen et al., 1983; Campbell and Brennen, 1985b; Johnson et al., 1990; Holyoake,
2011; Ottino and Khakhar, 2000; Khakhar et al., 1999; Savage and Lun, 1988; Pouliquen and
Renaut, 1996; Silbert et al., 2001; Ancey, 2001; Barbolini et al., 2005; Baran et al., 2006).
Most of the above experimental studies focus on fixed chutes. However, the chutes used in
blast furnaces are rotating; they may even rotate at such high rates that Coriolis and centrifu-
gal forces start to play a significant role, leading to prevailing flow patterns and particle dis-
tributions in a rotating chute which deviate considerably from those in a non-rotating chute.
Reliable measurements are essential to obtain a detailed understanding of such granular flows
in rotating chutes. Three-dimensional particle tracking velocimetry is a promising technique,
because it can potentially deliver with high spatial and temporal resolution the particle bed
height and 3D velocity profile. The bed height can also be obtained by using ultasonic height
sensors, but the resolution of such sensors is lower, while the experimental procedure is much
more labour intensive or expensive (a new experiment needs to be performed for each mea-
surement position, or a large number of sensors needs to be used). The velocity profile can
also be obtained by using particle image velocimetry, but usually only in two dimensions.
In this chapter we will focus on cross-validation of 3D-PTV technique, for applications of
granular flows in chutes rotating at significant rotation rates, against independent other tech-
niques. In chapter 2 we described in detail the PIV technique to measure the surface particle
velocity and the electronic ultrasonic sensor technique to measure the bed height in a rotating
chute. In this chapter, we add the three-dimensional PTV technique to obtain detailed infor-
mation about the tracks of the surface particles, leading to three dimensional surface particle
velocities and the bed height in the chute. This tracer particle tracking technique is chosen
Cross-validation of 3D-PTV for granular flows down rotating chutes 57
Table 4.1: Dimensionless Rossby (eq. 5.1) and Froude (eq. 5.2) numbers for simulations and
experiments.
Rotation rate Rossby number Froude number

Ω = 8 rpm 1.879 0.055
Ω = 16 rpm 0.939 0.218
Ω = 24 rpm 0.627 0.491
because we believe it is the most suitable measurement technique for collecting Lagrangian
data in granular flows. The PIV technique used in our previous work can give relatively high
temporal and spatial resolution flow measurements. However, the results obtained from this
technique are restricted to average velocity vector fields only, instead of real particle trajec-
tories. In other words, the PIV techniques provides an Eulerian description of the flow field,
with the mean velocity of a group of particles at different spatial positions. In contrast, PTV
is a flow measurement technique that identifies individual tracer particle displacements based
on a Lagrangian description. The advantage of PTV over PIV is that it provides truly local
velocity information for individual particles, but this leads to the drawback of difficulties with
identifying corresponding particles at different time steps. Therefore, PTV measurements are
restricted to flows with low seeding concentrations of tracer particles.
We perform our analysis on a system of monodisperse 3 mm spherical glass particles, flowing
down a rectangular plexiglass chute inclined at 30 degrees and rotating at various rotation
rates. In particular, we choose a particular range of rotation rates between 0 and 24 rotations
per minute because we expect a transition in the behaviour occurring within this range. The
dimensionless Rossby number and Froude numbers for the experimental setup, based on the
rotation rates and maximum velocity at the end of chute, are given in the Table 4.1. The mass
rate used is 3.2 kg/s. Note that in chapter 3 we showed a DEM validation, using experimental
results of surface velocity obtained by PIV and and bed height from an ultrasonic sensor,
using a lower mass rate of 1.6 kg/s.
We reiterate that the primary goal of this chapter is to assess the ability of the 3D-PTV
technique to provide surface information such as surface velocity of particles and particle
bed height in the rotating chute. Besides the advantages mentioned above, 3D-PTV provides
additional information compared to PIV, namely the third component (i.e. the depthwise)
velocity in the chute.
As a secondary goal, we will use the results of this work to further validate our discrete ele-
ment method (DEM) (Shirsath et al., 2014b). DEM simulations are becoming more and more
popular because they can resolve the granular flow behaviour on an individual particle level
(Cundall and Strack, 1979). DEM has been successfully used for simulation of many indus-
58 Chapter 4
trial granular flow systems such as flows in rotating drums (Ristow, 1996; Wightman et al.,
1998), fluidized beds (Di Maio et al., 2009; Kaneko et al., 1999) and hopper charging and
discharging flows (Nguyen et al., 2009; Yu and Saxén, 2010, 2011a). Recently, a number of
papers focusing on DEM simulations of granular flows in the blast furnace charging process
have appeared (Mio et al., 2008, 2009, 2010, 2012; Ho et al., 2009; Zhou et al., 2011; Yu
and Saxén, 2010, 2011a; Yu, 2013; Yu and Saxén, 2013; Wu et al., 2013; Bhattacharya and
McCarthy, 2014; Akashi et al., 2008; Zhang et al., 2014). Given its popularity, it is crucial
that the DEM method is properly validated against precise experiments, including cases in
which the process equipment is rotating.
The chapter is organized as follows. In section 2 we present the experimental set-up and its
procedure. In section 3, the measurement techniques, focusing on PTV, are presented. In
section 4, a change in particle introduction in the simulation model is explained. In section
5, we present our post-processing methods to obtain data from the numerical model which
are similar in spirit to the experimental measurements. In section 6, we present the exper-
imental results, including a repeatability study of the PTV experiments for bed height and
streamwise surface particle velocity. We compare the bed height from PTV with independent
measurements using an ultrasonic height sensor, and compare the surface velocity from PTV
with independent measurements using PIV. In section 7, we validate our DEM simulations by
comparing with the experimental findings of bed height, streamwise velocity and spanwise
velocity for different rotation rates of the chute. We end with our conclusions.
4.2 Experimental setup and procedure

In this section we describe the hardware of the 3D-PTV system, consisting of the rotating
table facility, camera system and illumination facility. We outline experimental settings and
the experimental protocol followed in the experiments. Note that details on the electronic
height sensor technique and PIV technique have already been presented in section 2.2.
Experimental setup
Figure 4.1 shows a schematic representation of the side view of the experimental setup. The
chute is made of plexiglass for optical accessibility. The experimental equipment includes
a hopper for storage of particles and a collection tank. A vacuum pump is attached to the
top of the hopper to recycle the granular material from the collection tank. The mass rate
is measured through a dynamic weighing scale, and surface information such as particle bed
height and the surface particle velocity field is measured through PTV, PIV and an electronic
ultrasonic sensor.
To minimise rotational flow of the granular material inside the hopper prior to deposition on
the chute, the rotation axis passes through the centre of the hopper. The bottom end of the
PTV camera
PTV camera
PTV camera
Hopper
Support stand
Plexiglass chute
Collection tank
Weighing scale
Rotating table
and acquisition
hardware
Figure 4.1: Schematic drawing of the experimental PTV setup, side view.
hopper can be fitted with different mouthpieces in order to generate different flow rates at the
exit of the hopper.
The granular particles are deposited onto a rectangular plexiglas chute with transparent side-
walls through which the moving granular material can be photographed. The interior of the
chute is 1 m in length, 10 cm in height, and 8 cm in width. The bottom wall of the chute is
white to achieve a better contrast in the photographs for detecting tracer particles. The whole
chute is fixed at its bottom to a strong metal plate of 1 cm thickness to minimise vibrations
caused bye the particles poured from the hopper mouth to the chute surface at the inlet. The
inclination angle of the chute (defined with respect to the horizontal) is adjustable between 0
and 70 degrees. At the end of the chute, a collection tank receives the granular particles. This
60 Chapter 4
Figure 4.2: The experimental setup with rotating table facility at the Fluid Dynamics Labo-
ratory (Department of Applied Physics, Tu/e ) and PTV cameras.
tank is placed on a dynamic weighing scale to measure the exit flow rate.
The whole setup is mounted on a rotating table, so that the flow is measured in the non-inertial
frame of reference. All the hardware, constituting the focusing system and the measurement
system, is located on the top of the table or surface below it as shown in figure 4.2. The
equipment is remotely controlled from an adjacent room for safety precautions during the
rotating experiments. Details of the rotating table and camera system are described in the
next subsections.
Rotating table
The rotating table at the Fluid Dynamics Laboratory (Department of Applied Physics, TU/e)
used in this research is designed to support heavy constructions upto the weight of 1000 kg.
Furthermore, the table has the ability to spin at a constant angular velocity in the range of 0.01
rad/s to 10 rad/s, with an accuracy of 0.005 Ω. Since the PIV and PTV measurements are
highly sensitive for camera movements during the experiment, a stable rotational operation
is required. Del Castello and Clercx (2013) and Van Bokhoven et al. (2009) employed this
rotating table facility for rotating turbulence studies. They did several experiments and tests
to determine the exact stability and accuracy of the table. First of all, the maximum misalign-
ment of the table (i.e. the maximum vertical displacement) is 7.5 μm at a rotation rate of 0.01
rad/s. Moreover, the residual angular acceleration is below 10−3 s−2 for rotation rates up to
10 rad/s. Finally, the standard deviation of the coordinate fluctuations is below 0.026 pixels
at a rotation rate of 10 rad/s, and thus the produced vibrations do not have influence on the
accuracy of the PIV and PTV measurements.
Camera system
Three high speed cameras are situated above the chute, used to capture the images that are
used for the optical measurement techniques particle image velocimetry (PIV) and particle
tracking velocimetry (PTV). PIV is used to measure the surface velocity of the granular flow,
whereas PTV is used to measure the trajectories of the individual particles along the length
of the chute. For PIV only one camera is needed to capture images of the granular flow. It
is positioned perpendicular above the center of the chute at a distance of 1.6 m (the middle
camera). For PTV three cameras are needed, one of which has the same position as for PIV,
while the two other cameras are placed at an up- and downstream position, respectively, and
are slightly tilted, see figure 4.1. From the three different angles the position of particles
can be determined and subsequently the trajectory of individual particles. A 28 mm lens
is used to cover the entire length of the chute for all three cameras. The three cameras are
synchronized so that images are captured at the same time from all cameras. The high speed
cameras (Photron FastcamX-1024PCI) have an internal memory of 12 GB RAM to save up
to 60 000 images, which is equal to a saving time of roughly 30 seconds when using a frame
rate of 2000 frames per second and a resolution of 128 x 1024 pixels. This saving time of 30
seconds is sufficient to execute a single experiment, since the maximum experimental time is
9 seconds (total weight granular material is 30 kg; flow rate is 3.2 kg/s). To be able to save
the captured images and to run the PIV/PTV software, a computer system is needed. After
each experiment, the images are transferred from the internal memory of the cameras to this
computer system. From there on, the images can be processed using the specific PIV and
PTV software.
62 Chapter 4
Illumination system
The high frame rate of the cameras implies a very short exposure time. Therefore, the illu-
mination system has to be strong and homogeneous enough for the cameras to see the light
reflected by the granular particles, including the tracer particles, in every part of the measure-
ment field. The LED light is supplied by Ledlight Group, Germany, and is using a LED light
power supply of 95 W max (Vout : 30-350 V DC, Lout : 350mA-constant current, power factor
correction (PFC): 0.98.) Four LED lights are used to illuminate the whole chute. Each LED
Light contains 72 small lights (CrystalLedTM 72 LED). Two lights are fixed at each side of
the chute, taking care that they do not block the field of view of the cameras, as shown in
figure 4.2.
Tracer particles
To properly visualize the granular material in the rotating chute for 3D-PTV it has to be
seeded with tracer particles. The physical properties of these particles should be close to the
properties of the flowing granular particles to guarantee that they properly represent the flow
behaviour. In our experiments we used 3 mm blue spherical glass spheres, which have the
same collision properties as the other transparent glass particles. We mixed 1 kg of these
tracer particles into 30 kg of glass particles.
Calibration plate
In figure 4.3, the top view of calibration plate used for the experiments is shown (bottom
figure). The calibration plate has a thickness of 0.009 m and is 0.08 m wide and 1 m long,
which is sufficient to cover the entire area of the chute. The calibration plate contains 5 x 62
calibration points. The calibration points have a diameter of 4 mm and the distance from the
center of one calibration point to the next is 16 mm. Furthermore, three calibration points
have a larger diameter, to identify the center and orientation of the calibration plate during
processing of the calibration phase.
Experimental procedure
The following steps are used during the start-up of the experiments. First, the granular parti-
cles (3 mm glass spheres) are filled in the top hopper. The rotating table is accelerated until it
reaches the desired steady rotation rate. Then an experiment is started by opening the outlet
of the hopper by remote control, and the granular material flows in downward direction along
the length of the chute. At the end of the chute, the granular material is collected into the
collection tank. Measurements are made continuously until the hopper is depleted. The time
range during which steady-state flow conditions apply is determined after the experiment by
analysing the time dependence of mass, height and velocity measurements.
4.3 Particle tracking velocimetry technique

The 3D particle tracking velocimetry technique is a flexible non intrusive image analysis
technique for flow measurements. It was first introduced by Chang et al. (1984) and was
further developed by Racca and M (1988). This technique has a history of development
for more than a decade at the Institute of Geodesy and Photogrammetry (IGP) and at the
Institute for Hydrodynamics and Water Resources Management (IHW) both at the Swiss
Federal Institute (ETH) Zurich (Maas, 1991; Malik et al., 1993; Maas, 1995; Maas et al.,
2002).
In contrast to PIV, PTV is able to track individual particles in the flow and provides both the
Lagrangian and the Eulerian representation of the flow field (Willneff, 2002). To be able to
track individual particles instead of providing a global (Eulerian) flow field, PTV requires
three cameras to detect the position of the particle in the three dimensional domain. The
cameras capture images of the flow from different angles. From these images, it is possible
to determine the position of an individual particle and compute its trajectory. In PTV, it is
necessary to make a clear distinction between the particle which is being tracked and all
other particles. Therefore, tracer particles are introduced in the granular flow. Obviously,
the concentration of tracer particles should not be too high, otherwise it becomes difficult to
distinguish individual tracer particles, and the method becomes less accurate (Prasad, 2000).
The software code used for postprocessing is based on the ETH-code mentioned above, which
is modified and adapted for this setup by specialists of the Fluid Dynamics Laboratory at the
Department of Applied Physics (TU/e). The code forms the basis of the PTV system with
which the image sequences from the three cameras are processed and the 4D-coordinates {x,
y, z, t} of the tracer particles are reconstructed. The software code is freely available for non-
commercial use. The processing basically consists of three phases: calibration of the camera
system on a known target body, reconstruction of 3D-positions from image to object space,
and temporal tracking (Del Castello, 2010).
The calibration procedure adopted in our setup is explained in detail in the following sec-
tion. Once calibrated and synchronized, the camera system continuously records the tracer
positions. The image point coordinates of all tracer particles in the image are detected for all
cameras and all time instants using image processing and image analysis techniques. After
assignment of corresponding particle images from different views it is possible to determine
the 3D position in the object space by forward intersection. Using epipolar constraints, ho-
mologous image points can be detected and the 3D object coordinate of each particle can be
determined. In the last step, temporal matching is performed to determine trajectories of each
tracer particle. With the present setup, up to 400 particles per time step have been tracked on
average.
64 Chapter 4
Details of the algorithms used in our code can be found in the literature: for calibration and
3D positioning algorithms we refer to Maas et al. (1993); for the temporal tracking algorithm
to Malik et al. (1993); and for the latest developments of the tracking routine to Willneff
(2002, 2003) and Willneff et al. (2012). Below we will discuss the most important steps.
Calibration of the camera system

Calibration of the camera means determining the extrinsic and intrinsic parameters using a
set of image points with known world coordinates. A multi-planar calibration method is
used, in which multiple images are captured of the same planar calibration plate at different
heights within the experimental volume. The accuracy of the calibration is highly dependent
on the correct positioning of the plate (Del Castello, 2010). For this reason, special care has
been taken to accurately position the calibration plate throughout the measurement volume,
to minimize the disadvantage of the multi-planar calibration method. The larger the volume
covered by the calibration plate, the smaller are interpolation and extrapolation errors of the
calibration functions (Del Castello, 2010).
To create the calibration functions which correlate the pixel information of each camera to
the world coordinates, recordings of the calibration unit have been carried out with the same
orientation: the calibration plate plane is parallel to the chute base. Translation in the co-
ordinate direction perpendicular to the calibration plate plane is carried out. The recordings
of the calibration plate are registered in 4 different positions, at a height of 9 mm, 19 mm,
34 mm and 54 mm, with typical positioning error of less than 0.5 mm. With the informa-
tion of the pixel size of the camera sensors, and the diameter of the circular voids, 3rd order
polynomials relate the pixel information to the physical dimensions of the calibration plate.
Linear interpolations and extrapolations of the generated polynomials are extended to the
whole measurement volume.
Three-dimensional positioning of the particles

The second important phase of 3D-PTV processing is the three-dimensional positioning of
particles in space from the raw images. Figure 4.3 shows an example of the raw images
simultaneously generated by the three cameras.
Several processing steps are required to obtain the particle position from the captured images.
The first two steps are preprocessing steps, namely high pass filtering, and particle detection
and location of the particles. High pass filtering is necessary to remove non-uniformities
in the background illumination (Willneff, 2002). For the detected and located particles in
the preprocessing stage, the corresponding particles for all cameras are established. Since the
corresponding particles are detected by all cameras, subsequently the 3D coordinates of these
Camera 1
Camera 2
Camera 3
Top view of calibration plate
Figure 4.3: Raw images of granular flow in a chute rotating at 24 rpm, captured by 3 PTV
cameras at the same time instant. The group of tracer particles circled in the three images
represents the same particles as seen by the different cameras. The bottom figure shows the
top view of the calibration plate used in the experiments.
particles can be determined with the obtained calibration data from the first phase. From the
3D coordinates, the particles can be tracked in the object and image space.
Particle trajectory reconstruction

After obtaining the 3D coordinates of each particle, a particle trajectory is constructed by
comparing the positions of particles in consecutive image frames. For each frame, the soft-
ware tries to find an individual particle based on the position of that particular particle in the
previous frame(s). Later the displacement of an individual particle between two frames (a
single time step) is evaluated. In order to find such a displacement, at least two cameras are
needed. The trajectory of a particle that is detected by an individual camera can be imaged
as a two-dimensional path, also called epipolar lines. If a second camera detects the same
particle trajectory, a three-dimensional path can be determined from both two-dimensional
pathways. In PTV practice, three or even four cameras are used. The redundant information
is used to improve the accuracy of the method. In this research, only trajectories of parti-
66 Chapter 4
Figure 4.4: Snapshot of particle introduction at the inlet of the chute.
cles which are detected by all three cameras were used for further analysis. Furthermore,
because particle trajectories with just a few number of positions have a higher probability of
being false than trajectories with a large number of positions, we only used trajectories with
a mininum number of positions. This procedure has proven very useful to remove unreal-
istic particle trajectories. Elimination of particle trajectories comprising less than 10 spatial
positions has been found to be adequate in our work.

The results of this work will be used to further validate the discrete element method (DEM)
presented in chapter 3. A complete description of the simulation model and all settings can
be found in section 3.2. The only difference is the manner of introduction of the particles. In
chapter 2 we introduced the particles in a rectangular area with a bcc-lattice arrangement of

the particles, parallel to the chute bottom wall, where the initial velocity is set by the required
mass rate. We then found that the resulting bed height and particle velocity profiles were
independent of the exact manner of introducing the particles after a distance of approximately
0.2 m from the inlet. To reproduce the experiments somewhat more closely, in this chapter
we introduce the particles parallel to gravity, but still using the bcc-lattice arrangement for
convenience. Figure 4.4 shows a snapshot of introducing particles at the inlet in this manner.
We determined the first length wise position at which the bed height and surface flow velocity
match between experiment and simulation for a non-rotating chute at an angle of inclination
of 30 degrees, and subsequently placed the axis of rotation in our simulations at the same
distance from this location as in the experiments to ensure the same centrifugal forces are felt
by the particles.
4.5 Postprocessing of experimental and simulation data

Calculation of bed height from experimental PTV data
To obtain the particle bed height at different positions in the chute from the PTV data, a
postprocessing step is necessary. We assume that the bed surface can be defined as the height
at which the visible track density (in the height direction) is highest. This is better than using
an envelope around the top-most tracks, because such a measure would be dominated by
positions of stray particles. To determine the local track density, the chute volume is first
divided into computational cells. For each cell, the number of tracks that pass through it is
calculated, where tracks with a length of more than 10 positions were used. Subsequently, for
each column of cells (columns oriented in the y-direction of figure 2.1) the local maximum
of the density can be ascertained, and thus the height of the bed surface is known (within
the precision of a cell size). We divided the entire chute in 250 (length) × 16 (width) × 200
(height) = 800000 cells. This division was chosen to strike the best balance between a small
cell size in the height direction and a sufficiently large number of tracks per cell.
Figure 4.5 (blue line) shows an example of the bed height obtained in this manner for a
non-rotating chute, as a function of streamwise position and averaged over the width of the
chute. Note that the measurements are still somewhat noisy. Therefore, a Gaussian filter of
width 0.07 m is used to smoothen the curve as shown in figure 4.5 (green circles). This filter
width is large enough to significantly smoothen the data, yet small enough to still resolve the
dependence of bed height on streamwise position.
Calculation of bed height in simulations

In the DEM simulations, the bed height is calculated via the center of mass (COM) height. In
an ideal situation of homogenous packing, the bed height is exactly twice the center of mass
68 Chapter 4
0.035
PTV raw data
PTV smooth data
0.03
0.025
Bed height (m)
0.02
0.015
0.01
0.005
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length (m)
Figure 4.5: Experimental particle bed height defined by the highest track density from PTV,
for a non-rotating and rotating chute. The blue line represents the raw data from the post-
processing,the blue circles the smoothed bed height profile. Note that some raw data at the
inlet (negative values of the length position) was included in the smoothing, leading to an
apparent overshoot of the smooth data near length position 0.
Figure 4.6: For comparison with PTV data, the bed height in our simulations is calculated as
twice the center of mass (COM) height minus half a particle diameter.
height. However, we note that the bed height obtained from PTV measurements is basically
a result of tracking the centers of particles. To enable a fair comparison between experiments
and simulations, we should therefore base the bed height on the average height of the centers
of the surface particles. As explained in figure 4.6, the bed height (h) can be obtained by
subtracting half a particle diameter from twice the COM height.
Calculation of experimental surface velocity using PIV

We used particle image velocimetry (PIV) to determine the average two-dimensional velocity
field of the visible surface particles. Pairs of images are postprocessed using the commercial
sofware package DaVis 8.0.3. A multipass algorithm was used with an initial interrogation
area size of 32x32 pixels and a final non-overlapping interrogation area size of 16x16 pix-
els, which yields an approximated displacement error of O(0.1) pixels. Due to the very high
seeding density (particles per interrogation area) there were practically no outliers. Any re-
maining outliers were removed with a standard median filter. For rotating chute flow, there
may be a significant sideways motion and parts of the chute may fall “dry”, meaning that
only stray particles are present in some regions. Masking was performed on these dry re-
gions, meaning that no cross correlation analysis was applied in the masked region of the
chute. After post-processing, a time-averaged velocity field was obtained from a sequence of
instantaneous velocity fields obtained over a time interval during which the flow was steady.
We found that the error in this velocity field is 0.2%, which is relatively small.
Calculation of experimental surface velocity using PTV

Similar to PIV, it is possible to determine the surface velocity profile using the PTV technique.
From the known positions of tracer particle tracks at different time instances and the time
difference in between two frames, we can derive the velocity profile. Although PTV is a
3D technique, in these dense granular flows it is limited to tracking surface particles (those
visible from the top). Thus the results (for the streamwise and spanwise components) are
expected to match the surface particle velocity obtained from the PIV technique.
The procedure for obtaining Eulerian velocity profiles from the Lagrangian PTV tracks is as
follows. 1) The tracks are smoothed using a (Matlab) loess filter to avoid irregularities. 2)
The domain is divided into 250 (length) × 9 (width) columns. Coordinates of tracks inside
each column are identified. 3) The distance between each consecutive point of a track in a
column is found, and multiplied by the frame rate to obtain the velocity of that track section.
4) The average velocity for each column is calculated to obtain the surface velocity profile
as a function of spanwise and streamwise position. Note that the 3D-PTV technique also
delivers information about the depthwise velocity component of the surface particles, which
cannot be obtained from the PIV experiments.
Calculation of surface velocity in simulations

To enable a direct comparison between experimental and simulated surface velocity, a two
dimensional particle velocity field is calculated also in our simulations. It is important to only
include those particles that would be optically visible from above the chute, because in the
70 Chapter 4
experiments the camera was mounted perpendicularly above the chute at a large distance of
1.5 m. Specifically, in our simulations, the surface velocity is calculated by time-averaging
the velocity of the topmost 6 particles in each streamwise and spanwise column of 5 mm
× 6 mm. The choice for this number of particles was made because at the given particle
diameter of 3 mm, it is expected that at most 6 particles will be visible from above the chute in
each of the computational columns. We have checked the dependence of the surface velocity
measurements on the number of topmost particles used, and found only a negligible influence
for particle numbers ranging from 3 to 12.
4.6 Experimental results and discussion
Repeatability of PTV experiments

To investigate the repeatability of the PTV measurements, as well as to rule out possible
effects of wear and tear of the glass particles, some experiments have been performed twice:
once at the start of our experimental series, and once near the end. Figure 4.7(a) shows the
average streamwise surface particle velocity along the centerline of the chute as a function
of the streamwise position, for a non-rotating and rotating chute. Figure 4.7(b) shows the
average bed height along the centerline of the chute as a function of the streamwise position,
for a non-rotating chute. The results show that the repeatability of our PTV experiments is
satisfactory.
Comparison of bed height using PTV and the ultrasonic sensor

Figure 4.8 shows the bed height as measured in a non-rotating chute using PTV measurements
(blue symbols) and the electronic ultrasonic sensor (red symbols) for a non-rotating chute
inclined at 30 degrees. These experiments were performed with an interval of 4 months.
As explained in section 4.5, the bed height calculated in the PTV experiments is based on
the maximum density of visible tracks in the height direction, whereas the ultrasonic sensor
measures the bed height over a circular area of 28 mm in diameter. We note that in the initial
section, up to 0.2 m from the chute entrance, the electronic height sensor measurements were
somewhat affected by the noise (sound) generated by particles dropping from the hopper
onto the chute surface. This may explain the small deviations in the initial section. In the
remaining section of the chute, the electronic height sensor measurements follow the PTV
measurements quite closely, with an off-set of the order of one particle radius (1.5 mm). This
bias may be explained by the fact that sound waves scatter off the top of the particles, while
PTV is tracking the centers of particles, as explained in figure 4.6. Overall, the agreement
between the two independent experimental techniques is very satisfactory.
2.5
2 PTV Expt 1 @ = 0 rpm

PTV Expt 2 @ = 0 rpm
1.5
0.5
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length(m)
(a)
0.035
0.03 PTV Expt 1 @ = 24 rpm
0.025
Bed height (m)
0.02
0.015
0.01
0.005
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length (m)
(b)
Figure 4.7: Repeatability of measurements from PTV for a non-rotating and rotating chute in-
clined at 30 degrees. These results have been obtained at the start and end of our experimental
series, respectively.(a) streamwise surface velocity, and (b) bed height along the center-line
of the chute.
72 Chapter 4
0.035
PTV
Electronic Ultrasonic Sensor
0.03
0.025
Bed height (m)
0.02
0.015
0.01
0.005
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length (m)
Figure 4.8: Bed height obtained from PTV measurements (blue line) and electronic height
sensor measurements (green symbol) for a non-rotating chute (0 rpm) inclined at 30 degrees.
2.5
2
PTV expt @ Ω = 0 rpm
PIV expt @ Ω = 0 rpm
PTV expt @ Ω = 24 rpm
1.5
PIV expt @ Ω = 24 rpm
0.5
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length(m)
Figure 4.9: Averaged streamwise surface particle velocity along the center-line obtained from
PTV and PIV measurements for a non-rotating chute (0 rpm) and a rotating chute (24 rpm)
inclined at 30 degrees. Note that the PIV measurements stop at 0.66 m for the rotating case
because of masking (see main text).
Comparison of streamwise velocity using PTV and PIV

Figure 4.9 shows the averaged streamwise surface velocity along the center-line of the chute
using PTV measurements (blue line) and PIV measurements (green line), for a non-rotating
chute and a chute rotating at 24 rpm. As observed in our previous work (Shirsath et al., 2012,
2013), the granular flow decelerates roughly in the first half of the chute and accelerates in the
second half (both relative to the non-rotating case), which we attributed to a changing balance
between Coriolis and centrifugal forces. For the purpose of this chapter, it is encouraging to
see that the two independent measurement techniques, one based on Lagrangian tracking of
tracer particles and the other based on spatiotemporal correlations of all particles, are in good
mutual agreement.
In the next section we will show more detailed comparisons between PTV and PIV measure-
ments, but also include our DEM simulation results to validate the simulation method.
4.7 Validation of DEM using experimental measurements

Having performed a cross-validation of independent experimental measurements, we now
use these measurements to validate our discrete element model.
Figure 4.10 shows from a top view: the experiment, the PIV velocity field, a collection of
PTV particle tracks, and the simulation, respectively, for monodisperse glass particles in a
chute inclined at 30 degrees for a non-rotating chute (left) and a chute rotating at 24 rpm
(right). All results are in qualitative agreement with each other. The major effect of chute
rotation is a sideways deflection of the particle stream due to Coriolis forces present in the
frame of reference co-rotating with the chute. In the next sections we will present a detailed
quantitative comparison.
Bed surface height

Figure 4.11 shows the full bed surface as a collection of PTV particle tracks (left) and the full
bed surface obtained from our DEM simulations (right) for non-rotating (top) and rotating
(bottom) chutes.
Figure 4.12 provides a more detailed comparison, showing the bed surface height as a func-
tion of position along the length of the chute, for a chute inclined at a fixed angle of 30
degrees and rotating at various rotation rates. The bed height is averaged over 5 mm wide
sections of the chute, for three different width-wise positions in the chute (on the left, centre
and right-hand side). The simulation results (lines) are compared with PTV experimental re-
sults (symbols). For the non-rotating chute, the bed height continuously decreases along the
length of the chute, and the results are nearly indistinguishable for the different width-wise
positions. As the rotation rate of the chute increases, the bed height increases on the right
74 Chapter 4
Figure 4.10: Top view of the chute. Flow is from top to bottom. RAW is a snapshot from
the central camera. PIV is the surface particle velocity field determined by particle image
velocimetry. PTV is a collection of individual particle tracks used for particle tracking ve-
locimetry; different tracks are indicated by different colors. DEM is snapshot from the sim-
ulation, where particles are color coded by their velocity (blue = low velocity, red = high
velocity). (a) Non-rotating chute. (b) Chute rotating at 24 rpm.
side of the chute and decreases on the left side. Moreover, at higher rotation rates we observe
a maximum in the height as a function of streamwise position at the right side of the chute.
Some deviations between simulation and experiments are observed at the right side of the
chute (blue lines) at higher rotation rates, consistent with our previous observation that the
precise manner of particle introduction is important for the first 0.2 m. Given the experi-
mental error of approximately half a particle diameter, these simulation results are in good
agreement with the experimental measurements.
Streamwise surface particle velocity

Figure 4.13 shows the streamwise surface particle velocity as a function of width-wise po-
sition at four different streamwise positions, for rotation rates 0 rpm (non-rotating chute), 8
(a) (b)
(c) (d)
Figure 4.11: Surface plot for monodisperse particle flows down a chute inclined at 30 deg
and rotation rates of 0 rpm (a and b) and 24 rpm (c and d). Left (a and c): particle tracks from
PTV experiments. Right (b and d): bed surface from DEM simulations. Note that for clarity
we have chosen a view in which the flow is from right to left.
rpm, 16 rpm and 24 rpm. The streamwise positions are at z = 0.0 m, 0.2 m, 0.4 m and 0.6 m.
The simulation results (lines) are compared with experimental results of PIV (filled symbols)
and PTV (open symbols). Generally we find good agreement between the simulations and
both sets of experimental results. Some deviations between the PIV and PTV measurements
are visible at higher rotation rates. This is caused by the sideways motion of the granular par-
ticles, leaving one side of the chute empty with occasional stray particles. These stray particle
regions are masked out from our PIV analysis, while the stray particles are still processed up
by our PTV analysis. The observed deviations therefore correspond to velocities of very low
numbers of particles.
Spanwise surface particle velocity

Figure 4.14 shows the spanwise surface particle velocity along the width of the chute at
different cross sections in the length of the chute. The simulation results are compared with
76 Chapter 4
Figure 4.12: Bed height as a function of stream-wise position (along the length of the chute)
for three different width-wise positions in a chute inclined at 30 deg rotating with rates of 0
rpm, 8 rpm, 16 rpm and 24 rpm, respectively. Small circles are PTV experiments and lines
are simulations.
= 0 rpm = 8 rpm
3 3
2 2
1 1
0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
= 16 rpm = 24 rpm
3 3
2 2
1 1
0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
Chute width(m)
Figure 4.13: Streamwise particle velocity along the width of the chute inclined at 30 deg and
rotation rate of 0 rpm, 8 rpm, 16 rpm and 24 rpm. Filled symbols are PIV measurements,
open symbols are PTV measurements, and lines are simulation measurements.
= 0 rpm = 8 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
= 16 rpm = 24 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08

Chute width (m)
Figure 4.14: Spanwise particle velocity along the width of the chute inclined at 30 deg and
rotation rate of 0 rpm, 8 rpm, 16 rpm and 24 rpm. Open symbols are PTV measurements,
closed symbols are PIV measurements, and lines are simulation measurements. Different
colors represent different streamwise positions: 0.0 m (blue), 0.2 m (red), 0.4 m (green) and
0.6 m (black).
experimental results for different rotation rates of the chute at an inclination of 30 degrees.
We observe that the magnitude of the spanwise velocity first increases and then decreases
for consecutive streamwise positions. This corresponds to the process of sideways motion
induced by Coriolis forces, which is finally stopped by the compaction of the granular flow
against the side wall. The maximum spanwise velocity increases with increasing rotation
rate. Some deviations between simulation and experiments are observed at the start of the
chute (for z ≤ 0.2 m) because the way of introducing particles in the simulations is different
from experiments. In all other cases, a rather good agreement is found between experiments
and simulations.
Depthwise surface particle velocity

Figure 4.15 shows the depthwise surface particle velocity along the length of the chute for
different spanwise sections of the chute, as obtained from the PTV experiments. Note that for
the non-rotating chute, the depthwise velocity is high near the inlet of the chute and gradually
decreases to zero. For rotating chutes, there is a marked difference between relatively small
depthwise velocities measured at the left side of the chute (blue lines) and relatively large
depthwise velocities at the start of the right side of the chute (red lines). This is consistent
with the observed bed height profiles from our simulations and PTV tracks. We confirmed
that the slope of the bed height is consistent with the ratio of streamwise to depthwise velocity
78 Chapter 4
0.1 0.1
0.05 0.05
Depthwise surface particle velocity (m/s)
0 0
−0.05 x = 0.02 m −0.05
−0.1 x = 0.04 m −0.1
x = 0.06 m
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Ω = 16 rpm Ω = 24 rpm
0.1 0.1
0.05 0.05
0 0
−0.05 −0.05
−0.1 −0.1
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8

Chute length (m)
Figure 4.15: Depth wise particle velocity along the width of the chute inclined at 30 degree
and rotation rate of 0 rpm, 8 rpm, 16 rpm and 24 rpm. Lines are PTV measurements at the
left side (blue), center (green) and right side (red) of the chute.
(not shown). This confirms that the particles move parallel to the bed surface.
Depth dependent solid volume fraction

We can use the simulations to gain more insight into properties which are difficult to ac-
cess experimentally. One example is the depth dependent solid volume fraction. Figure 4.16
shows the contour plots of solid volume fraction for slices at three different spanwise posi-
tions. The simulation results show that the bottom part of the granular flow in a chute is close
to random close packing, but the solids volume fraction decreases towards the free surface,
with the largest gradient near the free surface. For a non-rotating chute, the solid fraction
along the center line is slightly different from the side positions. For the case of a rotat-
ing chute, the particles move sideways which results in an increase in solid volume fraction
(compaction) on the right side of the chute. Compared to our previous work (Shirsath et al.,
2014b) which was using a lower mass flow rate, the volume of highly packed particles is
larger, also relatively speaking. This can be explained by the increase in hydrostatic particle
pressure associated with a deeper particle bed.
4.8 Conclusions
In this chapter, a three-dimensional particle tracking technique (3D-PTV) was applied for the
first time to granular flow through a rotating chute. The technique was used to study bed
= 0 rpm
0.08
x = 0.015 m
0.5
0.06 0.1
0.2
0.
0.4
3 0.4
0.04
0
0
0.2
0.02 0.1
0.2
0.3
0.4
0.5
0.5 0.1
0.2
0.3
0.4 0.1
0.2
0.5 0.3
0.4 0.5
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0.08 0.6
x = 0.045 m
0.06
5
0.1
0.
0.2
0.4
0.
0.4
3
0.04
0
0
0.2
0.02 0.6 0.1
0.2
0.3
0.4 0
5
0.5
0.
0.1
0.2
0.3
0.4 0.5 0.1
0.2
0.3
0.4
0.6
0.6
0.6 0.5
0.6 0.6
0.6
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0.08 0.6
x = 0.065
0.5
0.06
0.1
0.4
0.2
0.
0.4
3
0.04
0
0.2
0.02 0.1
0.2
0.3
0.4
0.5
0.5 0.1
0.2
0.3
0.4 0.1
0.2
0.5 0.3
0.4 0.5
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
(a)
Height (m)
= 24 rpm
0.08 0.6
x = 0.015 m
0.5
0.06
0.4
0.1
0
0.
0.
02.4
3
0.1
0.04 0 0.1
0.2
0.3
0.2
0.3
0.1
0.2
0.4 0.5 0.4 0.3
0.5 0.4 0.2
0.02
0.5
0.5
0 0.6
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0.08 0.6
x = 0.045 m
0.3
0.06
0.5
0.1
0.4
0.2 0
0.04 0
.5
0
0.4
0.6 0.4
0.1 0
0.2
0.02 0.3 0.2 0.1
0.6 0.5
0.4 0.3 0.2 0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0.08 2
0.6
0.
x = 0.065 m
0.1
0.5
0.06
0.3
0.4
0.04
0.4
0.2
0.02 0
5
0.
0 0.4 0.3 0.2 0.1

0 0.5
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Chute length(m)
(b)
Figure 4.16: Contour plots of average solid volume fraction along the depth of the chute for
an inclination angle of 30 degrees and rotation rate of 0 rpm (a), and 24 rpm (b). Averages
are taken over slices with a width of 0.025m located at the left, center and right side of the
chute.
80 Chapter 4
height and surface particle velocities. The 3D-PTV experiments show very good repeatabil-
ity. A cross-validation of the particle bed height and the two dimensional velocity field was
performed by comparing with independent results obtained using an electronic ultrasonic
height sensor and particle image velocimetry. All experimental results are in good mutual
agreement.
The 3D-PTV technique has several advantages over other existing techniques. The advan-
tages relative to the use of an electronic height sensor for the measurement of the bed height
are, first, a better spatial resolution of the bed height profile and, second, a great reduction of
the number of necessary experiments (unless a large number of height sensors is used). The
advantage of 3D-PTV relative to the use of particle image velocimetry (PIV) is the measure-
ment of the third (depthwise) component of the surface particle velocity. An overall advan-
tage of 3D-PTV is that it can provide the measurements of particle bed height and surface
particle velocity field within one and the same experiment.
We have used the experimental results to further validate our discrete element model. We
have shown that the simulation model is capable of predicting the evolution of the granular
flow both qualitatively and quantatively.
Chapter
5
Simulation study of the effect of wall
roughness on the dynamics of
granular flows in rotating
semi-cylindrical chutes ∗
A discrete element model (DEM) is used to investigate the behavior of spherical particles
flowing down a semi-cylindrical inclined rotating chute. Such a system is of particular inter-
est for applications of charging of granular material to blast furnaces in the steel industry.
We present a validation of the DEM simulations by comparing with Particle Tracking Ve-
locimetry (PTV) experimental results of spherical glass particles flowing through a smooth
semi-cylindrical chute at different rotation rates of the chute. The DEM model predictions
agree well with the experimental results of surface velocity and particle bed height evolution.
Next the validated DEM model is used to investigate the influence of chute roughness on the
flow behaviour of monodisperse granular particles in rotating chutes. To emulate different
base roughnesses, a semi-cylindrical rough base is constructed out of a square close packing
of fixed spherical particles with a diameter equal to, smaller, or larger than the flowing par-
ticles. Finally, the DEM model is used to study segregation in a binary density mixture for
different degrees of roughness of the chute.
∗ This chapter is based on: Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. Simulation study
of the effect of wall roughness on the dynamics of granular flows in rotating semi-cylindrical chutes. Submitted to
AIChE Journal.
81
82 Chapter 5
5.1 Introduction
Granular flows down inclined planes are encountered in applications of industrial transport
of solid materials, as well as many naturally occurring phenomenon such as debris flows,
avalanches and landslides. The details of the flow are very sensitive to various parameters
such as geometry and roughness of the confining walls, shape of the grains, flow rate, and
coupling with the interstitial fluid.
In this chapter, we focus on the effect of wall roughness. Granular flows can be classified
into three categories based on the base roughness. The first is the flat-frictional base, in
which a relatively smooth planar surface has frictional interactions with the flowing particles.
The second is the bumpy base, which can be studied experimentally by glueing particles
onto the bottom surface (Pouliquen, 1999). The third is an erodible base in which a moving
layer of particles flows over a bed made up of static unconstrained particles (Kumaran and
Bharathraj, 2013). Experiments and simulations have shown that different flow regimes are
observed with different types of bases (Delannay et al., 2007; Thornton et al., 2012b). There
is shearing throughout the height for flow over a bumpy base, while the flow over a flat
frictional base consists of a thin mobile fluidized layer of particles at the bottom supporting a
plug flow above (Shirsath et al., 2014b). In the flow over an erodible bed, the flow is confined
to a thin layer of particles at the top of the pile. These qualitative differences in the flow
dynamics are not captured by the current phenomenological models.
There have been different approaches to attempt to understand and predict the flow dynamics
of granular materials, such as continuum theories based on kinetic theory of granular particles
and discrete particle simulations applied to granular flows. Before proceeding with granular
flows down rotating rough chutes, a short overview of literature on granular flows on inclined
planes without rotation is given.
Several authors have focused on flows on smooth frictional planes (Augenstein and Hogg,
1978; Brennen et al., 1983; Campbell and Brennen, 1985b; Johnson et al., 1990; Louge et al.,
1997). In particular, Campbell and Brennen (1985a) reported simulations of flows of granular
materials (two-dimensional flow of disks) down inclined chutes and compared the results
with existing experimental data. They found qualitative agreement with the velocity profiles
observed by Augenstein and Hogg (1978). They observed that the velocity shear profile
has the same shape as in liquid flow, except for a slip velocity at the wall which is much
higher than observed in liquid flow. The solid volume fraction is maximum at the center of
the flow, and reduced close to the bottom wall and close to the upper surface. The granular
temperature, which is a measure for the amount of velocity fluctuations, is found to be high
in the high-shear region next to the wall and low below the surface. Jenkins (1992) developed
the first theory for rapid granular flows interacting with a flat, frictional wall. Louge et al.
Effect of wall roughness on granular flows down rotating semi-cylindrical chutes 83
(1997) performed computer simulations to test the theory of Jenkins (1992) for the interaction
between a rapid granular flow of spheres and a flat, frictional wall. He examined the boundary
conditions that relate the shear stress and energy flux at the wall to the normal stress, slip
velocity, and fluctuation energy, and to the parameters that characterize a collision. He found
that while the theory captures the trends of the boundary conditions at low friction, it does
not anticipate their behavior at large friction.
Other authors have focused on flows on rough chutes (Drake, 1990; Savage and Hutter, 1991;
Ancey et al., 1996; Pouliquen and Renaut, 1996; Zheng and Hill, 1998; Daerr and Douady,
1999; Pouliquen, 1999; Dippel and Wolf, 1999; Hanes and Walton, 2000; Chevoir et al.,
2001; Silbert et al., 2001; Pouliquen and Forterre, 2002; Goujon et al., 2003; MiDia, 2004;
Thornton et al., 2012b). Dippel and Wolf (1999) found in their study of molecular dynamic
simulations of granular flow on a rough inclined plane in 2D, that the coefficient of restitution
has only a minor influence on the solid volume fraction as well as mean velocity of particles
in the steady state flow of granular materials. Augenstein and Hogg (1978) experimentally
measured the velocity profiles under a variety of conditions, and the effect of variables such
as as roughness and inclination of the surface, depth of flow and particle size has been eval-
uated quantitatively. They performed experiments with various sizes of particles glued to the
bottom of the chute. The flow of individual particles strongly depends on the nature of the
surface over which they are flowing. They found that when the glued particles are of the
same size as that of the flowing particles, the velocity profile exhibited zero slip at the wall.
On the other hand, when the glued particles are smaller than the flowing particles, the slip
increased towards that measured for a smooth surface. Similarly, Hanes and Walton (2000)
performed an experimental and discrete element model study of granular flows down bumpy
inclined chutes. They observed fully developed flows over a bumpy base for a range of in-
clinations. Moreover they reported that the characteristics of the base strongly influences the
flow regimes and flow dynamics. The experiments and simulations revealed relatively good
agreement for particle velocities near side-walls and on the surface. They observed that the
granular temperature is maximum near the bumpy base and decreases towards the surface.
GDR Midi (MiDia, 2004) proposed a phenomenological constitutive law for inclined plane
flows over uniform but rough chutes. Obviously, the particles at the bottom experience less
slip on rough planes, and their velocity at the wall is relatively small or even zero (no-slip).
Börzsönyi and Ecke (2006) performed experiments on the granular flows on rough inclined
channels with emphasis on high inclination angles. They characterized the granular flow by
measurements of the surface velocity, the average layer height, and the mean density of the
layer as a function of the hopper opening (i.e. mass rate), the plane inclination angle, and the
downstream distance of the flow. They observed that at low volume flow rates, a transition
was detected between dense and very dilute flow regimes. In addition it was observed that
84 Chapter 5
using a vacuum flow channel, air did not qualitatively or quantitatively modify the mean flow
velocities of the granular layer except for small changes in the very dilute like phase. Very
recently, Kumaran and Bharathraj (2013) studied the development of the flow of a granular
material down an inclined plane starting from rest as a function of the base roughness. Their
rough base was made of a random configuration of fixed spheres with diameter different from
the flowing particles, and the base roughness was decreased by decreasing the diameter of the
base particles. They confirmed that the flow development for the ordered and disordered flows
is very different. During the development of the disordered flow for the rougher base, there is
shearing throughout the height. During the development of the ordered flow for the smoother
base, there is a shear layer at the bottom and a plug region with no internal shearing above.
Despite the common occurrence of rotating chutes in industrial applications, only few studies
have investigated the effect of chute rotation on granular flows. The studies which have
been done are limited to smooth chutes (Mio et al., 2009; Shirsath et al., 2013, 2014b). In
the current work we will investigate the interaction between effects of base roughness and
effects of chute rotation. Our investigation will predominantly be based on DEM simulations,
but we start by validating the simulation method. We do this by a careful comparison with
experimental results for the case of a rotating semi-cylindrical smooth chute.
The experimental measurement methods should not influence the granular flows, i.e. they
should be non-intrusive. In the past, several non-intrusive measurement techniques have
been applied to granular flow systems, such as particle image velocimetry (PIV) (Lueptow
et al., 2000; Jain et al., 2002; Pudasaini et al., 2007; Shirsath et al., 2013), particle tracking
velocimetry (PTV) (Ottino and Khakhar, 2001; Pohlman et al., 2006; Pudasaini and Hutter,
2007; Shirsath et al., 2014a), magnetic particle tracking (MPT) (Buist et al., 2014), magnetic
resonance imaging (MRI)(Nakagawa et al., 1993; Cheng et al., 2006; Kawaguchi, 2010),
and positron emission particle tracking (PEPT) (Parker et al., 1997; Hoomans et al., 2001).
In our previous work (Shirsath et al., 2014a) we cross-validated results obtained from the
PTV technique against results obtained from the PIV technique for surface particle velocity
in both streamwise and spanwise direction, and against results from an ultrasonic sensor
for the particle bed height, finding good mutual agreement between the different methods.
Advantage of the PTV technique is that all measurements can be obtained from a single
experiment, whereas bed height measurements by an ultrasonic sensor at multiple locations
require a large number of experiments. Therefore, in the current work we apply the PTV
technique to investigate granular flows in a rotating semicylindrical chute.
The chutes used in blast furnaces are rotating; they may even rotate at such high rates that
Coriolis and centrifugal forces start to play a significant role, leading to prevailing flow pat-
terns and particle distributions in a rotating chute which deviate considerably from those in
a non-rotating chute. The key parameter that classifies the relative effect of the inertial and
Coriolis forces is the so-called Rossby number, defined as
va
Ro = , (5.1)
2ΩL cos φ
where va is the particulate phase velocity (within the co-rotating frame), Ω is the rotation rate
of the chute, L is a chute length and φ is the angle of inclination of the chute with respect to the
horizontal direction. This is an important dimensionless parameter also in geophysical flows
when comparing typical flow time scales vLa with relative rate of the Earth. When Ro 1, the
effects of rotation are unimportant and can be neglected, whereas Ro 1 signifies a system
which is strongly affected by Coriolis forces. The relative importance of the centrifugal force
compared to the gravitational force can be quantified by a (rotational) Froude number defined
as Mellmann (2001),
Ω2 L cos φ
Fr = . (5.2)
g
For the work described in this chapter we implemented a semicylindrical chute geometry
in the DEM model used previously for rectangular chutes (Shirsath et al., 2014b). We will
show that the DEM model shows good agreement with data obtained from PTV experiments.
The validated DEM model then is used to computationally study the effect of variation of
base roughness on the flow behaviour of monodisperse 3mm diameter particles in a rotating
rough chute. The base of the chute is comprised of a square packing arrangement of fixed
particles, and the diameter of the rough base particles is varied with respect to that of the
flowing particles in order to change the base roughness. A flat frictional base may be viewed
as the limiting situation in which the fixed particle diameter tends to zero. Specifically, for
the bumpy rough base, we choose particle diameters of 1.5mm, 3mm and 6mm, respectively.
We will also present a DEM simulation study of the influence of base roughness on the
segregation rate in a binary mixture of particles having the same diameter but different density
of 4000 kg/m3 and 900 kg/m3 .
This chapter is organised as follows. In section 2, the PTV experimental setup and technique
are discussed briefly. In section 3, a summary of the DEM model is given, including imple-
mentation details on the smooth and rough base and the parameter settings. A validation of
the DEM model against experimental data is presented in section 4. In section 5 and 6 we
present our base roughness study and segregation study, respectively. Finally, we give our
conclusions in section 7.
86 Chapter 5
-2ma(xva)
va
mag
Figure 5.1: Orientation of gravitational and Coriolis forces experienced by particles flowing
down a rotating inclined chute.
5.2 3D PTV measurements for rotating granular flows

Full details of experimental setup of the 3D-PTV system as applied to study granular flows
in a rotating semi-cylindrical chute can be found in chapter 4, where we applied the same
experimental technique to a rotating chute with a rectangular geometry.

Discrete element model
A complete description of the simulation model can be found in chapter 3. The rotation axis
and the directions of the Coriolis acceleration are all shown in figure 5.1.
Implementation of the smooth and rough wall chute

Two different types of bases, namely a smooth and a rough base, are used in the DEM model
simulations.
For the smooth base, almost the same contact model is used as for a particle-particle contact.
One difference is that the normal unit vector defining the collision contact orientation is
always pointing perpendicular to the local surface of the semi-cylinder (i.e. towards the
long symmetry axis of the corresponding cylinder). The other difference is that a higher
coefficient of friction is used for particle-wall contacts. This corresponds to the higher friction
experienced between the glass beads and a plexiglass wall.
The rough base is made up of spherical particles which are fixed in a square close packing in
a semi-cylindrical manner and which remain stationary in the co-rotating frame of reference.
In all cases, the rough base has been created by placing the particles against the smooth
semicylindrical wall. We use the same contact model between flowing and stationary base
particles as between two flowing particles, but again using a higher coefficient of friction. By
choosing different diameters for the stationary base particles, a variation in the base roughness
can be obtained. Three different stationary base particle diameters of 1.5 mm, 3 mm and 6
mm (0.5, 1 and 2 times the diameter of the flowing particles) are analyzed in detail. In order
to make fair comparison of different base roughnesses, the coefficient of friction between the
rough base particles and the flowing particles is made equal to friction between the smooth
base and flowing particles. A cross-section of the smooth chute and three arrangements of
rough base particles are shown in figure 5.2. We note that in all simulations smooth sidewalls
are present above the semicylindrical chute, keeping all particles inside the system until they
leave at the end of the chute.
Simulation and experimental settings

The physical properties of the spherical glass particles and the conditions for the simulations
and experiments are shown in Table 5.1. The resulting Rossby and Froude numbers are given
in Table 5.2.
In our DEM simulations the chute is initially empty, as is the case in the physical experiments.
Simulations are carried out for a constant mass flow rate at the inlet of the chute. We introduce
the particles at the top of chute near entrance in a rectangular feeding area with arrangement
of the particles in a bcc-lattice. We introduce the particles with a constant mass flow rate,
which is equal to the mass flow rate in the experiments.
Because in the experiments the particles are fed by a hopper, the details of particle intro-
duction are slightly different between experiments and simulations. We have shown before
(Shirsath et al., 2014b) that flow characteristics, such as surface particle velocity and particle
bed height, in a similar system (but using a rectangular chute) are sensitive to the details of
particle introduction, but only in the first 30 cm of the chute just after the sluice gate. More-
over, the first 10 cm is blocked from the field of view of the PTV cameras by the hopper.
88 Chapter 5
(a) (b)
(c) (d)
Figure 5.2: Cross-sectional view of the implementation of the smooth chute (a) and three
different rough base particle diameters: (b) Db = 1.5 mm, (c) Db = 3.0 mm, and (d) Db = 6
mm.
Therefore, here and in the following, we will define the lengthwise chute position z = 0 as
the point which is physically 10 cm from the intersection between the rotation axis and the
bottom of the chute, being aware that there may still be small differences between experi-
ments and simulations for z values up to 20 cm.
Time averages of the particle flow properties are calculated in the steady state. The meaning
of steady state in our simulations is that the mass flow rate at the exit is equal to the mass flow
rate at the inlet of the chute. With our mass rate of 3.2 kg/s, the total number of particles in
the chute reaches up to 55,000 for a smooth base, and 60,000, 75,000 and 90,000 for a rough
base particle diameter of Db = 1.5 mm, 3 mm and 6 mm, respectively. The simulations were
carried out on a single core of an Intel Xenon E5520 processor (at 2.27 GHz).
Computational measurements
For all computational measurements the chute domain is first divided in a number of compu-
tational cells. The grid size is chosen in such a way that a sufficient number of particles is
present in each cell, while ensuring that the particles in the sampling cell have a correlated
mean velocity. In the current work, the grid size is chosen as 28 cells in the width direc-
tion, 28 cells in the height direction, and 150 cells in the length direction, so that each fully
Table 5.1: Simulation and experimental settings.
Property Simulation Experimental

Length of chute 0.9 m 1.0 m
Diameter of chute 0.14 m 0.14 m
Inclination of chute 30 deg 30 deg
Rotation rate of chute 0, 4, 8, 16 rpm 0, 4, 8, 16 rpm
Mass flow rate 3.2 kg/s 3.2 kg/s
Particle type Spherical glass Spherical glass
Particle diameter 3 mm 3 mm
Particle density 2550 kg/m3 2550 kg/m3
Coefficient of friction μ pp = 0.10, μ pw = 0.22
Number of computational cells 28 x 28 x 150
Table 5.2: Rossby and Froude numbers for different chute rotation rates Ω. The Rossby
number is given for the smooth chute (Db = 0) and the roughest chute (Db = 6 mm). The
Froude number is independent of chute roughness.
Ω Ro (Db = 0) Ro (Db = 6 mm) Fr

4 rpm 4.5 2.96 0.012
8 rpm 2.3 1.49 0.049
16 rpm 1.1 0.74 0.19
90 Chapter 5
occupied cell contains on average around 6 particles. Note that we do not use a cylindrical
coordinate system to define the mesh, first because this would lead to a very low number of
particles per cell near the symmetry axis, and second, because the symmetry is broken by
gravity. Concerning the latter, the natural choice for the coordinate system is one in which
the long axis of the chute and the gravity direction form a single coordinate plane (in our case
in the zy plane), and the last coordinate (the x coordinate) is perpendicular to this plane .
To compare data from PTV and DEM for the smooth chute case, the particular postprocessing
needed for comparing bed height and surface particle velocity is described in detail in Shirsath
et al. (2014a). In brief, to obtain the simulated surface particle velocity, for each column of
computational cells the average velocity is determined of only the topmost particles which in
reality would be visible from above by the PTV cameras. The bed height in each column of
computational cells is estimated as twice the center-of-mass height of the particles, relative
to the chute bottom at the location of the column. Note that this approach is only well defined
for smooth chutes where the granular flow is plug-like and the surface of the granular medium
is relatively sharp.
In this work we also use the DEM simulations to gain insight in the effect of base roughness
on the granular temperature. Similar to the concept of temperature in statistical mechanics,
the granular temperature θ is a measure for the velocity fluctuations relative to the average
flow velocity. The granular temperature is a very important quantity in the kinetic theory of
granular flow (KTGF), determining the rate of particle collisions (Gidaspow, 1994; Nieuw-
land et al., 1996). DEM simulations such as used in present study can be used to verify basic
assumptions underlying the KTGF, and to validate the application of continuum models based
on KTGF to these types of flows.
We use the method described in Goldschmidt et al. (2001) to calculate the granular tempera-
ture. Note that the results are somewhat grid-size dependent: when the grid size is reduced,
the particle dynamics will become more homogeneous, leading to a lower granular tempera-
ture. On the other hand, when the grid size is increased, more spatial variations of the particle
dynamics will come into play, leading to a higher granular temperature. It is therefore impor-
tant to clearly define the optimal grid size when comparing different systems. In our case the
grid size is chosen to be 5 mm. This is a good compromise between having enough particles
within each grid cell for good statistics, and a small enough size to be able to measure the
flow features with relatively high resolution.
The granular temperature in the x-direction in each cell k is computed from:
1 Nk
Nk a∈∑
θk,x = (va,x − ūk,x )2 , (5.3)
cell k
where the sum runs over all particle a inside cell k, Nk is the total number of particles in cell
k, and ūk,x the average particle x-velocity in cell k:
1 Nk
Nk a∈∑
ūk,x = va,x . (5.4)
cell k
Granular temperatures for the y and z directions are calculated in a similar way. We will
investigate the anisotropy in granular temperature in the longitudinal direction relative to that
in the height direction (θz /θy ). As we will show, this anisotropy is significantly influenced
by the chute base roughness.
The overall granular temperature is obtained with contributions from all spatial directions:
1
θk ≡ (θk,x + θk,y + θk,z ). (5.5)
3
In this chapter, when an average granular temperature is reported, the average is weighted by
the number of particles in each cell. In this way cells that do not contain particles are not
taken into account in the average.
5.4 Validation of the discrete element model for a smooth

semicylindrical chute
Our DEM simulations for a semi-cylindrical chute are validated with well-defined laboratory
experiments with the settings defined in Table 5.1. Specifically, we will focus on the particle
bed height and streamwise and spanwise particle velocities at the surface of the granular flow
for different rotation rates at a fixed angle of inclination of 30 degrees.
Figure 5.3 shows a qualitative comparison of raw images, the collection of PTV particle
tracks, and snapshots from the DEM simulations. The major effect of chute rotation is a side-
ways deflection of the granular flow due to Coriolis forces present in the frame of reference
co-rotating with the chute. The colors in the DEM snapshots show how the particle velocity
increases from top to bottom of the chute and also increases with rotation of the chute. In the
following subsections we will make a more quantitative comparison.
Streamwise surface particle velocity as a function of width position

Figure 5.4 shows the streamwise surface particle velocity as a function of width-wise position
at four different streamwise positions, for rotation rates 0 rpm (non-rotating chute), 4 rpm, 8
rpm and 16 rpm. The streamwise positions are at z = 0.0 m, 0.2 m, 0.4 m and 0.6 m. The
simulation results (lines) are compared with PTV experimental results (symbols). At 0 rpm, it
can clearly be observed how the flow becomes narrower and the streamwise velocity increases
with increasing z-position. Furthermore, with increasing rotation rate, the particles are forced
92 Chapter 5
RAW PTV DEM RAW PTV DEM
(a) (b)
Figure 5.3: Top view snapshots of steady-state granular flows, flowing from top to bottom
through a chute inclined at 30 degrees. The chute is rotating with a rotation rate of (a) 0
rpm and (b) 16 rpm. The mass flow rate at the inlet of chute is equal to 3.2 kg/s in both
experiments and simulations. The first image is a raw image of the experiment, the second
image is a reconstructed image of PTV particle tracks for all times, and the last image is
a DEM simulation snapshot, where particles are color-coded according to their streamwise
velocity, from blue to red for low to high velocity.
= 0 rpm = 4 rpm
3 3
2 2
1 1
0 0
0 0.05 0.1 0 0.05 0.1
= 8 rpm = 16 rpm
3 3
2 2
1 1
0 0
0 0.05 0.1 0 0.05 0.1
Chute width (m)
Figure 5.4: Streamwise surface particle velocity as a function of width-wise position, at 4

different streamwise (z) positions. The chute is inclined at 30 degrees and rotating at a rate of
0 rpm, 4 rpm, 8 rpm and 16 rpm, respectively. Symbols represents experimental PTV results
and lines are simulation results.
to the right side of the chute due to Coriolis forces. Most importantly, the DEM simulation
predictions are in near-quantitative agreement with the experimental measurements. We find
a mismatch near the sides of the flow between our DEM simulations and PTV experiments.
This may be caused by the fact that fast particles, escaping to the side of the main flow, will
have long trajectories, while slower particles will re-enter the main flow after a relatively short
trajectory. This could lead to a bias towards higher velocities in our PTV velocity analysis.
Spanwise surface particle velocity as a function of width position

Figure 5.5 illustrates the spanwise surface particle velocity as a function of widthwise position
at the same streamwise positions as figure 5.4. At the top of the chute z = 0 m, an outward
velocity, relative to the center of the chute, can be observed. For the left side of the chute
this velocity is negative, and for the right side positive. Since the mouth piece of the hopper
is rectangular and positioned at the center of the chute, the particles tend to move outwards,
i.e. they splash. This splashing of the particles is different between DEM simulations and
experiments because the way of introducing the particles to the chute is similar but not exactly
the same. This leads to differences in the spanwise velocity, especially during the first 20 cm
of the particle stream. The current measurements show good agreement for z = 0.2 m and
beyond, confirming the insensitivity of particle introduction details for a semi-cylindrical
chute also.
94 Chapter 5
= 0 rpm = 4 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0.3 0.3
0.4 0.4
0 0.05 0.1 0 0.05 0.1
= 8 rpm = 16 rpm
0.1 0.1
0 0
0.1 0.1
0.2 0.2
0.3 0.3
0.4 0.4
0 0.05 0.1 0 0.05 0.1
Chute width (m)
Figure 5.5: Spanwise surface particle velocity as a function of width-wise position, at 4

different streamwise (z) positions. The chute is inclined at 30 degrees and rotating at a rate of
0 rpm, 4 rpm, 8 rpm and 16 rpm, respectively. Symbols represents experimental PTV results
and lines are simulation results.
In all cases we observe that the magnitude of the spanwise velocity first increases and then
decreases for consecutive streamwise positions. This corresponds to the process of sideways
motion induced by Coriolis forces. The maximum spanwise velocity increases with increas-
ing rotation rate.
Bed height
Figure 5.6 shows the bed height obtained from PTV measurement as a function of the stream-
wise position along the length of the chute. The bed height obtained from PTV is averaged
over a slice centered at widthwise positions x = 3.5 cm (at 1/4th of the chute diameter) and
x = 7 cm (at the centerline), respectively, using a slice width of one computational cell, i.e. 5
mm. The simulation results (lines) are compared with experimental results (symbols). For the
non-rotating chute, the bed height continuously decreases along the length of chute at both
width-wise positions. As the rotation rate of the chute is increased, the bed height increases
on the right side of the chute and decreases on half-width position of the chute. Moreover, at
higher rotation rates we observe a maximum in the height as a function of streamwise position
at the side of the chute. These observations are again in agreement with the sideways mo-
tion of the granular flow caused by Coriolis forces. Some deviations between simulations and
experiments are observed at the side of the chute (blue lines) at higher rotation rates in the ini-
tial 20 cm, which is due to the difference in particle inlet conditions, as discussed previously.
for two width-wise positions centered at x = 3.5 cm (blue) and x = 7 cm (red), respectively,
for rotation rates 0 rpm, 4 rpm, 8 rpm and 16 rpm. Symbols represents experimental results
and lines are simulations results.
Beyond 20 cm, the agreement between simulation and experiments is very satisfactory.
5.5 Influence of base roughness on monodisperse granular flows

We now use the validated DEM simulation method to investigate the influence of base rough-
ness on the flow behaviour of monodisperse granular materials.
Figure 5.7 shows snapshots for the monodisperse granular flow through the chute inclined
at 30 degree and rotation rates of 0 rpm (top) and 16 rpm (bottom), comparing the smooth
chute with rough chutes with increasing degree of roughness. Figure (a)-(d) shows that, as
the roughness of the chute increases, the particle velocity in the chute decreases down the
chute because of the higher resistance offered by the base. At the same time, the flow also
becomes more dilute because of the increased perturbations offered by the stronger undulat-
ing base. Because the mass rate is fixed, both effects leads to a significant increase in height
and broadening of the particle stream. Figure (e)-(h) shows that the chute rotation does not
essentially change this picture.
Bed height along the length of the chute

Figure 5.8 quantifies the increase in averaged height of the particle bed in the chute along
the length of the chute for different base roughnesses. The average height is maximum at the
96 Chapter 5
Ω = 0 rpm
(a) (b) (c) (d)
Ω = 16 rpm
(e) (f ) (g) (h)
Figure 5.7: Snapshots of DEM simulations of monodisperse granular flows in a chute inclined
at 30 degrees for a rotation rate of 0 rpm (top) and 16rpm (bottom). Particles are colored
according to their streamwise velocities. The blue color indicates the chute wall. (a) smooth
wall, (b) base Db = 1.5 mm, (c) Db = 3 mm, and (d) Db = 6 mm. Similarly for (e)-(h).
Figure 5.8: Averaged bed height along the length of the chute inclined at 30 degrees for
different base roughnesses and different rotation rates.
Figure 5.9: Averaged streamwise particle velocity as a function of the position along the
chute for different base roughnesses (see legend) at 0 rpm (left) and 16 rpm (right).
inlet of the chute and then slowly decreases along the remainder of the length of the chute for
a smooth chute. As the roughness increases the bed height in the chute increases as compared
to smooth chute. As the rotation rate increases, the averaged bed height in the chute increases,
most notably at the highest rotation rate where the centrifugal forces are strongest. This is in
agreement with the Froude number of 0.19 at the highest rotation rate at the end of the chute.
Streamwise particle velocity as a function of the position along the chute

Figure 5.9 shows the cross-sectional averaged streamwise particle velocity along the length of
the chute for different base roughness and rotation rates. As the base roughness is increased,
we observe that the average velocity decreases along the length of the chute. Flowing parti-
98 Chapter 5
(a) z = 0.0 m (b) z = 0.5 m (c) z = 0.9 m

3 2.5 3
2
2 2
1.5
1
1 1
0.5
0 0 0
0 0.05 0.1 0 0.05 0.1 0 0.05 0.1
(d) z = 0.0 m (e) z = 0.5 m (f) z = 0.9 m

3 2.5 3
2
2 2
1.5
1
1 1
0.5
0 0 0
0 0.05 0.1 0 0.05 0.1 0 0.05 0.1
Chute width (m)
Figure 5.10: Averaged streamwise particle velocity as a function of the lateral (width) posi-
tion for the non-rotating case (top) and 16 rpm (bottom) for different base roughnesses (see
legend). The flow profiles are measured at three different positions along the length of the
chute.
cles close to a fixed rough base experience a higher resistance, which may enhanced by an
increased entanglement and ordering of the flowing material near a rigid surface Pouliquen
and Renaut (1996); Pouliquen (1999). In this respect it is important to note that a smooth
base always produces an ever accelerating flow, whereas in sufficiently rough and sufficiently
long chutes, the averaged velocity will reach a steady state. Fig. 5.9 shows that our chutes are
not long enough to reach such a steady state, even for the highest amount of roughness.
When the chute is rotating, there is a small increase in velocity. Generally, the increase is
of the order of 10% at the highest rotation rate of 16 rpm. The relative increase compared
to the nonrotating case is different for different base roughness: the slope of the curves in
figure 5.9 is observed to become more equal at high rotation rate. This means that the average
acceleration of the particles in the granular flow becomes less sensitive to the base roughness
with increasing rotation rate.
Streamwise particle velocity as a function of width position

Figure 5.10 shows a striking effect of base roughness on the streamwise surface particle
velocity as a function of the lateral (width) position. At three different positions along the
length of chute, the surface streamwise velocity against the width of the chute is plotted for
the non-rotating and rotating chute. Flows over the smooth base develop a plug region in the
Figure 5.11: Averaged streamwise particle velocity as a function of height in a nonrotating

chute (0 rpm) for different base roughness.
centre of the chute, whereas flows over the rough base develop a more gradual and continuous
change in velocity. The latter is caused by the development of a shear layer in which the
velocity varies linearly with distance from the bottom wall Holyoake and McElwaine (2012),
in combination with the curvature of the semi-cylindrical base. The effect is not yet apparent
near the particle inlet at z = 0.0 m, but is strong halfway and at the end of the chute. For the
rotating chute, a similar trend in the development of the surface velocity profile is observed,
with the expected lateral motion of particles.
Streamwise velocity as a function of height in the chute (velocity shear profiles)
Figures 5.11 and 5.12 show the streamwise velocity of particles as a function of height in the
chute for different streamwise positions in the chute for the nonrotating and rotating cases,
respectively. These are better known as the velocity shear profiles. The shear profiles were
measured at the center plane of the chute and averaged over a width of 2 computational cells
(1 cm). In the case of the nonrotating chute, we clearly see the effect of base roughness on
the shear profile. As the roughness increases, all velocities are diminished, where the largest
difference is observed in the slip velocity of the particles close to the base of the chute. At
larger streamwise positions, the slip velocity increases, except for the largest roughness of
6 mm where the slip velocity remains close to zero. For the rough base cases, the shear
layer thickness (where the velocity increases most significantly) is of the order of 3 cm or 10
particle diameters.
100 Chapter 5
Figure 5.12: Averaged streamwise particle velocity as a function of height in a chute rotating
at 16 rpm for different base roughness.
Figure 5.13: Averaged granular temperature of particles along the length of the chute inclined
at 30 degrees for different base roughnesses.
For a rotating chute, the shear profiles are similar to those in a nonrotating chute at the be-
ginning of the flow, but start to differ considerably further downstream. Despite the observed
increase in velocity magnitudes, the thickness of the shear boundary layer (for the rough base
cases) does not increase noticeably.
Granular temperature as a function of the position along the chute

Figure 5.13 shows the time-averaged granular temperature along the length of the chute for
different base roughnesses under different rotation rates. The granular temperature represents
not only the fluctuation intensity between the granular particles but also represents the energy
dissipation due to colliding particles. High intensity fluctuations represent a high collision
rate and large energy dissipation.
For a smooth chute, the granular temperature is observed to decrease immediately from the
initial value caused by dropping of the particles onto the chute surface. This is consistent with
the observed plug-flow profile where relative velocities between particles are suppressed.
On the contrary, for rough base chutes, the granular temperature increases with increasing
streamwise position along the chute and is maximum at the exit.
The granular temperature is significantly influenced by the base roughness. Figure 5.13 shows
that the average granular temperature increases from 0.002 to 0.03 m2 /s2 at the end of the
chute with increasing base roughness. A similar behaviour was also observed by Forterre and
Pouliquen (2001) in granular flows down inclined non-rotating rough chutes.
The rotation of the chute also influences the granular temperature, especially for the larger
roughnesses of 3 mm and 6 mm, where the maximum is significantly increased at the end of
the rough base chute as compared to the non-rotating case. The increase in granular temper-
ature could be caused by centrifugal forces on the particles resulting in an effective lift from
the base and larger streamwise velocities, and therefore more violent collisions with the fixed
base particles.
Distribution and anisotropy of granular temperature
We will now look in more detail at the distribution of the granular temperature over the chute.
Figures 5.14 and 5.15 shows the distribution of granular temperature in the vertical center
plane of the chute inclined at 30 degree and rotation rates of 0 rpm and 16 rpm for different
base roughnesses of the chute. We observe that the granular temperature is highest near the
bottom of the chute and decreases in the direction of the surface of the bed (indicated by
the thick red line), in agreement with observations made by Hanes and Walton (2000). The
magnitude of the granular temperature strongly increases with increasing base roughness, as
expected. For non-smooth chutes, we find that chute rotation has an amplifying effect on
granular temperature. This is most clear in the second half of the chute, where the high-
temperature regions reach closer to the bed surface in rotating chutes than in non-rotating
chutes.
In thermalized equilibrium systems the velocity fluctuations are isotropic, i.e. the same in
all directions. However, we are dealing with a non-equilibrium dissipative system here, and
anisotropies in the granular temperature may develop. Note that this is of high interest to the
development and validation of kinetic theories of granular flow, which often assume that the
velocity fluctuations are characterized by a single, scalar, granular temperature. See, however,
also Sela and Goldhirsch (1998), who predicted the granular temperature anisotropy to be
largest near the bottom of the chute.
102 Chapter 5
Db = 0 mm 3
x 10
0.06 6
0.04 4
0.02 2
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Db = 1.5 mm
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
L
Height (m)
Db = 3 mm
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Db = 6 mm
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length (m)
Figure 5.14: Contour plots for granular temperature of particles along the vertical center plane
of the chute inclined at 30 degrees and no rotation (0 rpm) for different base roughnesses.
Note the different color scales. The thick red line indicates the bed height of the particles in
the center plane.
Effect of wall roughness on granular flows down rotating semi-cylindrical chutes 103
Db = 0 mm −3
x 10
0.06 5
4
0.04
3
2
0.02
1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Db = 1.5 mm
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
L
Db = 3 mm
Height (m)
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Db = 6 mm
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Chute length (m)
Figure 5.15: Contour plots for granular temperature of particles along the vertical center
plane of the chute inclined at 30 degrees and a rotation rate of 16 rpm for different base
roughnesses. Note the different color scales. The thick red line indicates the bed height of
the particles in the center plane.
104 Chapter 5
Db = 0 mm
0.05
0.04
1.1
0.03
1
0.02
0.9
0.01
0 0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Db = 1.5 mm
0.05
0.04
1.1
0.03
1
0.02
0.9
0.01
0 0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Figure 5.16: Contour plots for longitudinal-to-height anisotropy of the granular temperature
(θz /θy ) along the vertical center plane of the chute inclined at 30 degree and no rotation (0
rpm) for different base roughnesses. Note that no measurements are possible at the height of
the fixed base particles (white areas at the bottom).
Db = 0 mm
0.05
0.04
1.1
0.03
1
0.02
0.9
0.01
0 0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Db = 1.5 mm
0.05
0.04
1.1
0.03
1
0.02
0.9
0.01
0 0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Figure 5.17: Contour plots for longitudinal-to-height anisotropy of the granular temperature
(θz /θy ) along the vertical center plane of the chute inclined at 30 degree and a rotation rate of
16 rpm for different base roughnesses. Note that no measurements are possible at the height
of the fixed base particles (white areas at the bottom).
106 Chapter 5
Figures 5.16 and 5.17 shows the distribution of the anisotropy in granular temperature (θz /θy )
in the vertical center plane of the chute for different base roughnesses. This measures the
ratio of velocity fluctuations in the stream direction relative to velocity fluctuations in the
height direction. For smooth chutes and rough chutes with a roughness of db = 1.5 mm we
find that this ratio is generally very close to unity, meaning that the amount of anisotropy is
low, except possibly very close to the chute base. For chutes with higher roughness we find
somewhat more anisotropy, with streamwise velocity fluctuations generally being larger than
the vertical velocity fluctuations by 10 to 20 percent. For the smooth chute, chute rotation
leads to a depletion of particles along the center plane, and no accurate measurements can be
made beyond 0.4 m. For the rough chutes, measurements can be made along the full length
of the chute. We find that chute rotation has a relatively small effect on the anisotropy in
granular temperature.
In summary, we find that chute rotation leads to higher granular temperatures, especially in
the second half of the chute. The anisotropy is generally low for smooth chutes and of the
order of 10-20% for rough chutes. The anisotropy is not influenced by chute rotation.
5.6 Influence of base roughness on segregation in binary density flows

We now turn our attention to the influence of bottom roughness and chute rotation on seg-
regation of particles in a binary mixture of high density (4000 kg/m3 ) and low density (900
kg/m3 ) particles. All simulation parameters (except density) are as given in Table 5.1. The
two different types of particles are introduced at the inlet of the chute in a randomly mixed
fashion with a particle number ratio of 1:1 and a constant mass flow rate of 3.2 kg/s.
Figure 5.18 shows snapshots of the binary mixture flowing through a chute inclined at 30
degree and rotation rates of 0 rpm and 16 rpm, respectively, where particles are colored
according to their density (green for high density, red for low density). For a smooth chute,
the segregation is very low. With increasing base roughness, the segregation visibly increases,
with the low density particles preferring a position closer to the surface of bed, and away
from the rough base. Already from the snapshots we can observe that chute rotation has
only a minor effect on the segregation rate. We will quantify this more fully in the following
subsections.
Center of mass height along the length of the chute

Figure 5.19 (top row) shows the center of mass height of the high density (symbols) and
low density (lines) particles along the length of the chute, for different base roughness and
for the nonrotating case and rotation at 16 rpm. For non-rotating chute as the particles flow
downwards along the chute, they segregate into a small density top layer and large density
Ω = 0 rpm
(a) (b) (c) (d)
Ω = 16 rpm
(e) (f ) (g) (h)
Figure 5.18: Snapshots of DEM simulations of binary mixture granular flows in a chute
inclined at 30 degrees for a rotation rate of 0 rpm (top) and 16 rpm (bottom). Particles are
colored according to their density: green for 4000 kg/m3 and red for 900 kg/m3 . The blue
color indicates the chute wall. (a) smooth wall, (b) base db = 1.5 mm, (c) db = 3 mm, and (d)
db = 6 mm. Similarly for (e)-(h).
108 Chapter 5
Figure 5.19: Influence of base rougness on density segregation. Top row: center of mass
height along the length of the chute for a rotation rate of 0 rpm (left) and 16 rpm (right).
The symbols are for large density particles and lines for low density particles. Bottom row:
difference in center of mass height along the length of the chute for a rotation rate of 0 rpm
(left) and 16 rpm (right).
bottom layer. The extent of segregation is very low for a smooth base because there are not
enough perturbations for the buoyancy to have an effect during the time needed to flow down
the chute. As the roughness increases, the degree of segregation increases because, first,
the perturbations are stronger, and second, the streamwise velocity becomes lower which
gives the particles more time to segregate. For the rotating chute, there is not a significant
change in segregation along the length of the chute. This is better quantified by looking at
the difference of center of mass height, as shown in the bottom row of figure 5.19. This plot
clearly shows that the segregation is nearly zero for a smooth chute, increases with increasing
base roughness, but seems to saturate when the base roughness is equal to the particle size
or larger. We observe now that chute rotation does tend to increase the segregation rate,
but the increase is relatively small. Note that for the rotating cases the true segregation is
slightly larger than estimated here because the normal of the bed surface changes its direction
with increasing rotation rate. The true segregation would be the segregation measured here
times the inverse cosine of the angle between the bed surface normal and the y-axis defined in
Figure 5.1. However, because this angle is evolving along the width and streamwise position,
this is not a practical manner of measuring the segregation. In any case, the influence of chute
rotation on the segregation is small.
Figure 5.20: Influence of base roughness on granular temperature in a binary density mixture.
Top row: averaged granular temperature along the length of the chute for a rotation rate of 0
rpm (left) and 16 rpm (right). The symbols are for large density particles and the lines for low
density particles. Bottom row: ratio of fluctuating kinetic energies, eq. 6.8, along the length
of the chute for a rotation rate of 0 rpm (left) and 16 rpm (right).
Granular temperature and equipartition of energy

Figure 5.20 (top row) shows the averaged granular temperature for small and large density
particles along the length of the chute for the nonrotating case and rotation at 16 rpm. As al-
ready observed for the monodisperse case, the granular temperature decreases for the smooth
chute, while it increases for all rough chutes with increasing length position.
The velocity fluctuations are consistently higher for the low density particles than for the
high density particles. This is reminiscent of the energy equipartition principle in statistical
mechanics, which states that close to equilibrium the kinetic energy of particle velocity fluc-
tuations is distributed equally among particles of different mass. Obviously our system is
not in equilibrium, but the departure from equilibrium may be quantified by investigating the
ratio α of fluctuating kinetic energies:
ml θl
α= , (5.6)
md θ d
where m is the mass of a particle, the subscript l applies to the light phase, and subscript d
to the heavy phase. Figure 5.20 (bottom row) shows that α is of the order of unity, meaning
that equipartition of energy approximately applies. However, the precise value of α ranges
from 0.6 to 0.8, depending in a non-trivial manner on position in the chute, base roughness,
and chute rotation rate. Note that the value of α lower than 1 is not related to the fact that
lighter particles tend to segregate because values in the same range are observed for the non-
segregating flow over a smooth chute.
110 Chapter 5
5.7 Conclusions
We have investigated the influence of base roughness, in combination with chute rotation,
on monodisperse and bidisperse (in density) granular flows through semi-cylindrical chutes.
Such chutes are commonly used in bell-less charging of blast furnaces in the steel industry.
We performed our study using DEM simulations, which we have first validated by comparing
with experimental PTV measurements of bed height and surface particle velocities.
We find that the base roughness has a strong influence on the bed height, average particle
velocity, and granular temperature of a monodisperse granular flows down inclined (rotating)
chutes. The slip velocity with the chute wall is much reduced for larger roughnesses, reaching
an essentially zero value when the roughness is twice the particle diameter. Rotation of
the chute leads, besides the obvious lateral motion (to the right) of the granular flow, to an
increase in bed height, an increase in average longitudinal velocity, a widening of the stream
velocity profile, and an increase in granular temperature, especially at downstream chute
positions.
The base roughness also has a strong influence on the segregation rate of a binary density
mixture. For a smooth (but frictional) chute, the segregation is essentially zero, it increases
strongly with increasing roughness, and finally saturates when the roughness is beyond a
particle diameter. Rotation of the chute tends to increase the segregation rate, but the increase
is relatively small. An analysis of the ratio of fluctuating kinetic energies of the light and
heavy phase shows that an equipartition of energy only approximately applies. The ratio
varies between 0.6 and 0.8, meaning that on average a larger fraction of the fluctuation kinetic
energy is residing in the heavy phase. Chute rotation tends to improve the equipartition of
energy. We anticipate that these observations will contribute to the future development and
validation of continuum models for application to granular flows in chutes.
Chapter
6
Simulation study of density
segregation in rotating rectangular
chutes
This chapter presents a numerical study based on the discrete element method (DEM) to
investigate segregation behaviour in a binary mixture of particles of equal diameter (3 mm)
but different density (ratio 3 or 6), flowing down an inclined rotating rectangular chute. The
distribution of radial positions of light and heavy particles on the burden surface plane are
theoretically calculated based on the exit positions and velocities of the particles, which gives
an indication of segregation patterns occurring on the burden surface. The observations
are correlated with measurements of the segregation rate and granular temperature. The
effects of chute rotation, chute inclination, particle feeding mass rate, particle density ratio,
roughness of the chute, and the insertion of a vertical wall along the length of the chute are
investigated. We find that segregation is enhanced by using steeper inclination angles, lower
feeding mass rates, and the use of rough chute walls or rough inserts. The main effect of
chute rotation is a widening of the burden distribution.
6.1 Introduction
The radial distribution of the mixture of raw granular materials (pellets, sinters and coke) on
the top of the particle bed in the blast furnace (i.e. the burden distribution) significantly affects
the distribution of gas flow, chemical reaction, and heat and mass transfer in the blast furnace.
A great effort has been spent on improving the production of steel by optimizing the blast
furnace operation. The charging of raw materials into the blast furnace is the critical operation
for the production of good-quality steel. These granular materials are charged from the top of
111
112 Chapter 6
a hopper onto the burden surface through an inclined rotating chute. The traditional rotating
chute has two rotational degrees of freedom, one vertical and one horizontal. A relatively
high rotation speed is given around the verical axis. The inclination of the chute is controlled
by setting the orientation around the horizontal axis. In the usual mode of operation, the
inclination angle is fixed or varying at a much slower rate than the rotation around the vertical
axis. With modifications of the charging mechanism (bell-less type charging), the flexibility
and diversity has been enhanced greatly. At the same time, the raw materials have different
shapes, sizes and densities. The tendency of these materials is to segregate when they flow
down through the chute and get distributed onto the burden surface. Different chute angles
and rotation rates produce different burden deposition patterns, with different segregation
patterns, which results in an inhomogeneous voidage distribution across the particle bed at the
top of a furnace. The inhomogeneous voidage distribution is closely related to the utilization
of gas and ultimately the fuel rate in the furnace. Thus, it is necessary to consider in detail
particle segregation during the charging process, which makes the system complicated to
understand. In addition, the coke particle layer collapses during ore charging because of the
large density difference between them (Ho et al., 2009). The coke particles are about three
times less heavy than the ore particles. Such phenomena make burden distribution even more
difficult to predict. Thus, it is important to understand and control the particle segregation
caused by the size and density difference taking place on the rotating chute as well as on the
burden surface of the blast furnace.
Segregation is unique to granular materials and has no parallel in fluids and, moreover, is an
important phenomenon concerning the flow of granular materials. Segregation of granular
materials has been a topic of intense investigations, but also for industrial frustration (Brown,
1939; Williams, 1976; Mullin, 2002; Ottino and Khakhar, 2000). A small change in particle
properties can lead to segregation where particles of one type tend to segregate or leave the
main stream of the flow. The mechanism of granular segregation for size and density is differ-
ent, the pattern of segregation is very similar in both cases where large or light particles rise
on the top of the flowing layer, and the small or heavier particles go down. There are differ-
ent mechanisms of segregation which are of particular relevance to shearing flow (Standish,
1979). The main driving force for density segregation, when the particle size is the same but
the density is different, is the effective buoyancy force experienced by the particles.
In the past, lots of studies have been carried out from both an experimental and numerical
point of view on granular flows down an inclined chute and most of the work is limited to
non-rotating chutes. The literature work related to chute flow, size and density segregation
will be summarized below.
Dolgunin and Ukolov (1995) and Dolgunin et al. (1998) studied segregation in a chute flow
Simulation study of density segregation down rotating rectangular chutes 113
for both size and density. For size segregation they used a close range (6.6 - 7.0 mm smooth
steel balls) while the density segregation experiments were performed for a nearly 1:2 density
ratio of two different materials. The interesting result from their work is that they obtained
a non-monotonic concentration profile for different components. They also validated their
continuum mathematical model of segregation by obtaining very good agreement with ex-
periments. The work of Dolgunin et al. appears to be one of the first continuum models
of segregation involving a chute flow, for both size and density segregation. Khakhar et al.
(1999) and Ottino and Khakhar (2000) studied the density and size segregation in a chute
flow of cohesionless spherical particles by means of computations and theory based on the
transport equations for a mixture of nearly elastic particles. Khakhar et al. (1999) derived
expressions for the segregation flux due to pressure gradients and temperature gradients in a
granular chute flow. The effect of density segregation and size segregation was also analyzed.
The theory showed that density segregation drives heavier particles to sink to lower regions of
the chute, while size segregation was influenced by the particle inelasticity and inter-particle
friction.
The study of rotating granular flows is also of fundamental interest, with many challenges
from both an experimental and numerical point of view. At the same time, many studies
have focused on the falling trajectory and burden distribution of granular materials during the
charging process (Radhakrishnan and Maruthy Ram, 2001; Nag and Koranne, 2009; Zhang
et al., 2014; Bhattacharya and McCarthy, 2014; Sawley et al., 2011; Yu and Saxén, 2011b;
Yu and Saxen, 2012; Yu, 2013; Kajiwara et al., 1984). Nevertheless, the mechanisms govern-
ing the formation of burden profiles are still not fully understood. The novelty of the work
presented here is that we study the density distribution on the burden surface under differ-
ent operating conditions of the chute (rotation rate, inclination angle, particle feeding mass
rate), including geometrical modifications of the chute, which has not yet been reported in
the literature mentioned above.
We perform our study using the discrete element model (DEM). DEM is one of the most
useful tools for the analysis, visualization, and study of closure relations for granular flows.
Cundall and Strack (1979) first used DEM to model the behaviour of particles of soil under
shear. The strong point of the DEM method is that, once validated by direct comparison with
the experiments, it allows for calculating physical quantities that are difficult or impossible
to measure in experiments. In recent years, DEM has been increasingly used to simulate
the mixing and segregation of granular particles in various systems. For example, mixing of
granular materials in a rotating drum was studied by Kostek and Landowski (2011). Yang
(2006) investigated the density segregation in vertically vibrated binary mixtures using DEM
simulations. He observed that the heavy particles tended to gather around the central region
and the concentration of light particles was higher at the top compared to that of heavy ones.
114 Chapter 6
In practice, segregation problems can be solved by two types of measures; changing the
particles or changing the process (Ottino and Lueptow, 2008). Changing the particles may
involve balancing the differences in size and density. Changing the process may involve
geometrical changes such as insertion of walls, rough base or operational changes such as
varying the rotational speed and inclination of the chute. We found that very little is known
about the effect of chute insertions on mixing and segregation. Moreover, a literature search
yields no previous studies dealing with issues involving novel chute geometry conditions and
their effect and how their designs will affect segregation.
The objective of this chapter is to investigate particle segregation in a binary density mixture
of particles flowing through inclined rotating chutes, using the DEM model. The trajectories
of low and high density particles, after they leave the chute, are theoretically extended to
the burden surface plane. We will show that chute rotation has an effect on low and high
density particles segregation in the chute, as well as on the distribution on the burden plane.
This chapter is organized as follows. First, the DEM model is explained briefly, including
calculation of the radial distance to the rotation axis of the particles hitting the burden plane.
Then, different cases are studied in detail. In particular, we study the effect of inclination
of the chute, the effect of the particle feeding mass rate, the effect of density ratio of the
particles, the effect of the base and side wall roughness, and finally the effect of insertion of
an additional wall into the chute. We summarize our conclusions.

Granular flow down an inclined rotating chute
The simulations of granular flow through the rotating chute were conducted with the discrete
element model in a frame of reference which is co-rotating with the chute, as described in
detail in chapter 3. To model a rough-wall chute, immobile particles were placed in a regular
rectangular arrangement at the chute boundaries, as described in detail in chapter 5.
Trajectories from the chute exit to the burden surface

Inside the chute, the particles feel reaction forces from the bottom and side walls, which
counteract the graviational and Coriolis forces, leading to a compaction of the particle stream
and therefore an enhancement of the particle collisions. After the particles leave the chute
exit, these reaction forces disappear and the particles are in free fall, as indicated schemati-
cally in figure 6.1. Since in industrial blast furnaces the density of the gas is small compared
to the particle density, and the particles are fairly large, the gas-induced forces will be ig-
nored (Shirsath et al., 2014b). Moreover, it will be assumed that the motion of the freely
falling particles will not be influenced by particle collisions because during their free fall the

Y
v
v
H (m)
rex vr
ex
rcr
Radial distance (m)
Figure 6.1: Side view of the free fall of a binary granular flow after exiting the chute. In
this work we calculate the radial distance to the rotation axis where particles cross the burden
surface at a height H = 1.6 m below the intersection of the chute bottom wall with the rotation
axis.
number density of the particles will rapidly decay due to fast spreading of the granular flow.
Therefore, we will assume that the trajectory of each particle is determined by its position
and velocity at the time when it is exiting the chute.
After the particles have fallen over some height, they will hit the burden surface. The exact
impact position where each particle hits the burden surface can be calculated easily. Specifi-
cally, we will focus on the distribution of radial distances rcr from the axis of rotation where
granular particles will cross the burden surface at a distance (measured along the axis of ro-
tation) of H = 1.6 m below the intersection of the chute bottom wall with the rotation axis,
see figure 6.1. We choose this height H because a free falling height of a few times the chute
length is typical for the steady-state operation of a blast furnace.
The easiest way to calculate rcr is to first determine the vertical, radial and azimuthal compo-
nents of the position and velocity in the fixed laboratory frame at the time when the particle
116 Chapter 6
is exiting the chute, which we indicate by ζ , r and θ , respectively (defining the burden plane
at ζ = 0). These can be related to the coordinates in the chute-oriented co-rotating frame of
reference, indicated by x, y, z, as follows:
ζ ex = H + yex cos φ − zex sin φ (6.1)

ex 2 1/2
rex = (x ) + (yex )2 + (zex )2 − (ζ ex − H)2 (6.2)
vζ = y cos φ − vz sin φ
vex ex
(6.3)
1 ex ex ex ez ex 2
vr = ex
vx x + vex
y z sin φ cos φ + y sin φ + vz z cos φ + y sin θ cos φ
ex 2 ex
r
(6.4)
1 ex ex
vθ = v (z cos φ + yex sin φ ) − xex vex
z cos φ + vy sin φ
ex
+ Ωrex (6.5)
rex x
where superscipts ex are used for the particle position and velocity at the exit, and the exit
position (xex , yex , zex ) is measured relative to the position of the intersection point of the rota-
tion axis with the bottom plane of the chute. We leave out the expression for θ ex because it
will be irrelevant for our purpose of determining rcr . The term Ωrex is added to the azimuthal
velocity vθ because we need this velocity in the laboratory (inertial) frame of reference.
Knowing the exit position and velocities in the fixed laboratory frame, we continue to calcu-
late rcr . From the time of exit onwards, the particle will be in free fall motion with a height
described by ζ (t) = ζ ex + vζ t − 12 gt 2 after an elapsed time t. This allows us to calculate the
time t cr elapsed until the particle will intersect with the burden surface at ζ = 0, and finally
the radial distance rcr of the intersection point from the rotation axis:

2 1/2
vζ vζ 2ζ ex
t cr = + + , (6.6)
g g g
1/2
rcr = (rex + vr t cr )2 + (vθ t cr )2 . (6.7)
Simulation settings
The physical properties of the spherical glass particles and plastic particles, the chute di-
mensions, and the flow conditions in our DEM simulations are shown in Table 6.1. In the
simulation, the chute is initially empty, as is the case in the physical experiments. Simulations
are carried out for a constant particle feeding mass rate of 2 kg/s at the inlet of the chute. We
= 0 rpm = 32 rpm
Figure 6.2: Snapshots of a binary density granular flow down the chute on the burden surface,
for both a non-rotating chute (0 rpm) and rotating chute (32 rpm). For illustration purposes
particle trajectories outside the chute have been included here.
introduce the particles at the top of the chute near the entrance in a rectangular area with a
bcc-lattice arrangement of the particles. Typical snapshots are shown in figure 6.2.
6.3 Overview of cases

We will investigate the flow behaviour, with a focus on segregation, of a binary density mix-
ture of light and heavy particles along the length of the inclined chutes as well as the radial
distribution of the deposition of these particles on the burden surface plane. The simulations
are performed for both the non-rotating chute and the rotating chute (32 rpm). The results ob-
tained from the simulations are the center of mass height of the particles, segregation rate, the
granular temperature and energy equipartition, all as a function of distance along the length
of the chute. Additionally, we show the radial distribution of light and heavy particles on the
burden surface. We will investigate, successively, the following cases:
1. effect of the chute inclination angle,

118 Chapter 6
Table 6.1: Simulation settings (common base-case).
Property Value
Inclination angle of chute 30 degree
Particle types Spherical glass and spherical plastic
Density of glass particle 2700 kg/m3
Density of plastic particle 900 kg/m3
Coefficient of friction μ pp = μ pw = 0.3
2. effect of the particle feeding mass rate,
3. effect of the density ratio of the particles,
4. effect of the chute roughness and insertion of a separating wall in the chute.
We repeat that the common base-case simulation parameters used in the different cases are
given in Table 6.1.
6.4 Case 1: effect of the chute inclination angle

We compare two different inclination angles of the chute: 30 degrees and 40 degrees in-
clination with respect to the horizontal plane, respectively. All other parameters are kept
unchanged.
Center of mass height and segregation rate

Figure 6.3 shows the averaged center of mass heights (top row) and difference between these
center of mass heights (bottom row) of small density and large density particles along the
length of the chute. We find that an increase in inclination angle from 30 to 40 degrees leads
to a relatively small change in COM height. The bottom figure shows that the segregation
0.03 0.03
φ = 30 deg
φ = 40 deg
COM height (m)
0.02 0.02
0.01 0.01
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
−3 −3
x 10 x 10
Difference of COM heights (m)
5 5
4 4
3 3
2 2
1 1
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length(m)
Figure 6.3: Influence of the angle of inclination of the chute on density segregation. Top row:
center of mass height along the length of the chute for a rotation rate of 0 rpm (left) and 32
rpm (right). The symbols are for high density particles and lines for low density particles.
Bottom row: difference in the center of mass height along the length of the chute for a rotation
rate of 0 rpm (left) and 32 rpm (right).
(defined as the difference in COM height of small and large density particles) is slightly in-
creased for the steeper inclination angle. When the chute is rotating, the segregation increases
for both cases.

Figure 6.4 (top row) shows the averaged granular temperature for light and heavy particles
along the length of the chute. The granular temperature is minimum at the inlet of the chute
and maximum at the exit of the chute. The granular temperature increases with increasing
inclination angle. This is consistent with the fact that a larger inclination angle increases the
particle velocities in the chute, which makes the flow more dilute and allows for more violent
collisions between the particles. Chute rotation tends to increase the granular temperature of
both heavy and light phases.
The velocity fluctuations are consistently higher for the light particles than for the heavy
particles. As in the previous chapter, we quantify the equipartion of energy by the ratio α of
120 Chapter 6
0.02 0.02
φ = 30 deg
φ = 40 deg
0.015 0.015
θ (m2/s2)
0.01 0.01
0.005 0.005
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
1.1 1.1
1 1
α (−)
0.9 0.9
0.8 0.8
0.7 0.7
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length(m)
Figure 6.4: Influence of angle of inclination of the chute on granular temperature in a binary
density mixture. Top row: averaged granular temperature along the length of the chute for
a rotation rate of 0 rpm (left) and 32 rpm (right). The symbols are for high density particles
and the lines for low density particles. Bottom row: ratio of fluctuating kinetic energies α for
a rotation rate of 0 rpm (left) and 32 rpm (right).
fluctuating kinetic energies:

m l θl
α= , (6.8)
md θ d
where the subscript l applies to the light phase and subscript d to the heavy phase. Figure
6.4 (bottom row) shows that the value of α along the length of chute tends to be closer to
unity for the steeper inclination angle for the non-rotating chute, but that it becomes saturated
at a value below unity for a rotating chute. It is difficult to give a physical explanation, but
these observations will be important for validation of continuum models based on kinetic
theory of granular flows: if such models assume an equipartition of energy (α = 1), they will
possibly predict erroneous values for granular fluid properties which depend on the granular
temperature, such as the effective viscosity.
Radial burden distribution

Figure 6.5 shows the radial distribution of light and heavy particles. The first obvious obser-
vation is that the average radial position decreases with increasing chute angle from 30 to 40
15 15
φ = 30 deg
φ = 40 deg
10 10
P(r)
5 5
0 0
1.2 1.4 1.6 1.8 2 2.2 1.2 1.4 1.6 1.8 2 2.2
Radial distance (m)
Figure 6.5: Radial distribution of light and heavy particles on the burden surface plane from
the axis of rotation of the chute for different inclinations of the chute (see legend) at 0 rpm
(left) and 32 rpm (right). The symbols are for large density particles and the lines for low
density particles.
degrees. The second observation is that the segregation between heavy and light particles in
the chute is also visible in the burden distribution. When the chute is rotating, the average
radial position shifts to larger values. At the same time, the distributions also become wider,
which is most easily appreciated from the decrease in peak height in these normalized plots.
The average position of the burden distribution can be predicted by simply inserting the
center-of-mass position and center-of-mass velocity of the particles exiting the chute (for
each phase) in equations (6.1)-(6.7). Concerning the spread, a comparison between figures
6.5 and 6.4 (top row) shows that the width of the distribution is also positively correlated to
the granular temperature at the exit.
6.5 Case 2: effect of particle feeding mass rate

We compare two different particle feeding mass rates of 2 kg/s and 4 kg/s, respectively. All
other parameters are kept unchanged.

Figure 6.6 shows the center of mass heights (top row) and difference of center of mass heights
(bottom figure) of light and heavy particles along the length of the chute. As expected, the
average COM height increases when increasing the mass rate from 2 kg/s to 4 kg/s. An
important observation is that the segregation rate actually decreases with increasing mass
rate, as is most clear from the bottom figure. This is consistent with a decrease in granular
temperature, as we will show in the next subsection. For both mass rates, rotation of the chute
enhances the segregation in the chute.
122 Chapter 6
0.04 0.04
Q = 2 kg/s
Q = 4 kg/s
COM height (m)
0.03 0.03
0.02 0.02
0.01 0.01
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
−3 Ω = 0 rpm −3
x 10 x 10
5 5
4 4
3 3
2 2
1 1
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length(m)
Figure 6.6: Influence of particle feeding mass rate on density segregation. Top row: center of
mass height along the length of the chute for a rotation rate of 0 rpm (left) and 32 rpm (right).
The symbols are for high density particles and lines for low density particles. Bottom row:
difference in center of mass height along the length of the chute for a rotation rate of 0 rpm
(left) and 32 rpm (right).

Figure 6.7 (top row) shows the granular temperature for light and heavy particles along the
length of the chute. Clearly the granular temperature decreases with increase in mass flow
rate from 2 kg/s to 4 kg/s. On the other hand, the granular temperature increases strongly
with increase in rotation rate of the chute, especially at downstream positions. Figure 6.7
(bottom row) shows that without rotation the value of α along the length of the chute tends
to 1 sooner for lower mass rates. However, when the chute is also rotating, the value of α
saturates at a value of approximately 0.75 for both mass rates.

Figure 6.8 shows the radial distribution of particles. Without chute rotation, the effect of
increasing the mass rate is a clear narrowing of the width of the distribution. This is consistent
with the lower granular temperature observed at higher mass rates.
Chute rotation increases the average radial position of the burden distribution. This is caused
by the higher flow velocity at the exit of the chute. Again, the average radial position can
0.015 0.015
Q = 2 kg/s
Q = 4 kg/s
0.01 0.01
θ (m2/s2)
0.005 0.005
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
1.1 1.1
1 1
α (−)
0.9 0.9
0.8 0.8
0.7 0.7
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length(m)
Figure 6.7: Influence of mass flow rate on granular temperature in a binary density mixture.
Top row: averaged granular temperature along the length of the chute for a rotation rate of
0 rpm (left) and 32 rpm (right). The symbols are for large density particles and the lines for
low density particles. Bottom row: ratio of fluctuating kinetic energies for a rotation rate of
0 rpm (left) and 32 rpm (right).
be predicted on the basis of the center-of-mass position and center-of-mass velocity at the
chute exit. Chute rotation also increases the width of the distribution, but at the same time the
distributions for the particle feeding mass rates of 2 kg/s and 4 kg/s also become more similar.
These observations are consistent with the fact that the granular temperatures become higher
but also more similar under chute rotation.
6.6 Case 3: effect of density ratio of the particles

We compare two different density ratios of 3 (2700 and 900 kg/m3 ) and 6 (5400 and 900
kg/m3 ), respectively. All other operating parameters are kept unchanged.

Figure 6.9 shows the center of mass heights (top row) and difference of center of mass heights
(bottom figure) of light and heavy particles along the length of the chute. Because the mass
rate and the light particle mass are fixed, a higher density ratio leads to a lower number of
particles inserted per second, and therefore a lower bed height.
124 Chapter 6
15 15
Q = 2 kg/s
Q = 4 kg/s
10 10
P(r)
5 5
0 0
1.2 1.4 1.6 1.8 2 2.2 1.2 1.4 1.6 1.8 2 2.2
Radial distance (m)
Figure 6.8: Radial distribution of light and heavy particles on the burden surface plane from
the axis of rotation of chute for different inflow mass flow rate of the chute (see legend) at
0 rpm (left) and 32 rpm (right). The symbols are for large density particles and the lines for
low density particles.
Concentrating on the segregation rate, we find that a larger density ratio clearly leads to a
larger segregation rate. Chute rotation enhances the segregation rate, with a similar relative
enhancement for both density ratios.
Figure 6.10 (top row) shows the granular temperature for light and heavy particles along the
length of the chute. The evolution of the granular temperature with length position is very
sensitive to the density ratio of the particles. Most notably, the granular temperature of the
light particles is increasing when the density ratio is increased, despite the fact that actually
the mass of the light particles is not changed in this comparison. This could be caused by the
lower number of particles at fixed mass rate for a higher density ratio: the effect is then similar
to the increase in granular temperature of the light particles for a lower mass rate observed in
Figure 6.7. The granular temperature of the heavy phase seems to be less influenced by the
density ratio. Chute rotation tends to increase the granular temperature. Again, the granular
temperature of the light phase is more strongly influenced than the granular temperature of
the heavy phase.
Figure 6.10 (bottom row) shows that the value of α along the length of chute is very sensitive
of the density ratio, changing from values below 1 to values above 1 when increasing the
density ratio from 3 to 6. As observed before, chute rotation tends to decrease and at the
same time more or less stabilize the value of α. The value of α is less than unity for both
density ratios beyond z = 0.2m.
0.03 0.03
Center of mass height(m)
Density ratio = 3
Density ratio = 6
0.02 0.02
0.01 0.01
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Ω = 0 rpm
0.01 0.01
0.008 0.008
0.006 0.006
0.004 0.004
0.002 0.002
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length(m)
Figure 6.9: Influence of density ratio of heavy and light particles on density segregation. Top
row: center of mass height along the length of the chute for a rotation rate of 0 rpm (left) and
32 rpm (right). The symbols are for high density particles and lines for light density particles.
Bottom row: difference in center of mass height along the length of the chute for a rotation
rate of 0 rpm (left) and 32 rpm (right).

Figure 6.11 shows the radial distribution of particles. Without chute rotation, increasing the
density ratio is clearly leading to a stronger segregation in the average deposition distance,
with the light particles ending up further from the rotation axis and heavy particles ending up
closer to the rotation axis.
Chute rotation increases the average position and increases the width of the distribution.
Despite the increase in segregation inside the chute, the amount of segregation between heavy
and light phase in the burden deposition distribution remains similar to the non-rotating case.
6.7 Case 4: effect of chute roughness and insertion of a dividing wall in

the chute
Finally we compare three different chute designs, namely the default smooth chute, a rough
chute, and a rough chute with an additional vertical rough wall inserted along the length of the
chute, exactly at the central (symmetry) plane. The rough walls are made by fixing particles
126 Chapter 6
0.02 0.02
Density ratio = 3
Density ratio = 6
0.015 0.015
θ (m2/s2)
0.01 0.01
0.005 0.005
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
1.4 1.4
1.2 1.2
α (−)
1 1
0.8 0.8
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8

Chute length(m)
Figure 6.10: Influence of density ratio of light and heavy particles in the chute on density
segregation. Top row: averaged granular temperature along the length of the chute for a
rotation rate of 0 rpm (left) and 32 rpm (right). The symbols are for large density particles
and the lines for low density particles. Bottom row: ratio of fluctuating kinetic energies for a
rotation rate of 0 rpm (left) and 32 rpm (right).
10 10
Density ratio = 3
8 Density ratio = 6 8
6 6
P(r)
4 4
2 2
0 0
1.2 1.4 1.6 1.8 2 2.2 1.2 1.4 1.6 1.8 2 2.2
Radial distance (m)
Figure 6.11: Radial distribution of light and heavy particles on the burden surface plane
from the axis of rotation of chute for different density ratios of light and heavy particles (see
legend) at 0 rpm (left) and 32 rpm (right). The symbols are for large density particles and the
lines for low density particles.
with a diameter equal to the diameter of the flowing particles (3 mm) to the walls in a regular
(square) pattern. All other parameters are kept unchanged. Figure 6.12 shows the prevailing
flow patterns of binary density particles down these three designs of rotating chutes. The
top row shows snapshots for a non-rotating chute. Note how the light density particles are
moving away from the side walls and bottom walls for the rought chute. The bottom figure
shows snapshots for rotating chute simulations.

Figure 6.13 shows the center of mass heights (top row) and difference of center of mass
heights (bottom figure) of light and heavy particles along the length of the chute. Let us first
focus on the non-rotating case (left figures). Increasing the roughness of the chute leads to
a higher bed, with a larger amount of segregation between the particles. Inserting a rough
dividing wall has a dramatic effect, significantly increasing the segregation.
The same observations are made when the chute is also rotating. Comparing the bottom left
and right figures, we find that, remarkably, chute rotation does not lead to any additional
amount of segregation for a rough chute or chute with rough insert.

Figure 6.14 (top row) shows the granular temperature for low and high density particles along
the length of the chute. For the non-rotating chute, as expected, the granular temperature is
higher in the rough chute as compared to a smooth chute. The granular temperature in the
chute with the rough wall insert falls somewhere between those of the smooth wall and the
rough wall beyond z = 0.3 m. This is caused by the fact that the larger amount of avail-
able rough surface also leads to a lower average flow velocity with somewhat less violent
collisions.
Chute rotation leads to a higher granular temperature in both the smooth and the rough chute
(blue and red lines). In contrast, for the rough chute with inserted wall (green lines), the
granular temperature is very similar to that of the non-rotating chute. This correlates with
our observation in Figure 6.13 (bottom row, green lines) that also the segregation rate is
not influenced by chute rotation for the case of a rough chute with inserted wall. In this
case, the particles are divided in two smaller segmented chutes by a rough insert. Therefore
more particles will be in close contact with walls, and their velocities will consequently be
perturbed by the roughness of these walls. This effect is clearly dominating over the effect of
rotation.
Figure 6.14 (bottom row) shows that the value of α, measuring the amount of equipartition
of fluctuating kinetic energy, is very sensitive to the chute roughness and presence of inserts.
128 Chapter 6
Figure 6.12: Snapshots of binary density particles flowing down an inclined rotating chute,
for a rotation rate of (a-c) 0 rpm and (d-f) 32 rpm. Three different chute designs are shown:
(a) and (d) for a smooth chute, (b) and (e) for a rough wall chute, and (c) and (f) for a rough-
wall chute with a vertical insertion at the central (symmetry) plane. Green particles are heavy
particles, red particles are light particles, blue particles are fixed rough-wall particles.
0.05 0.05
smooth chute
rough chute
Center of mass(m)
0.04 0.04
insertion
0.03 0.03
0.02 0.02
0.01 0.01
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
0.01 0.01
0.008 0.008
0.006 0.006
0.004 0.004
0.002 0.002
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length(m)
Figure 6.13: Influence of roughness of the chute and insertion of a wall in the rough chute
on density segregation. Top row: center of mass height along the length of the chute for a
rotation rate of 0 rpm (left) and 32 rpm (right). The symbols are for high density particles
and lines for low density particles. Bottom row: difference in center of mass height along the
length of the chute for a rotation rate of 0 rpm (left) and 32 rpm (right).
Moreover, the effect of changing from a smooth chute (blue lines) to a rough chute (red lines)
is opposite with or without rotation. In all cases, the chute with a rough insert leads to an α
closest to unity, meaning the best amount of equipartition of energy across the different types
of particles.

Figure 6.15 shows that the chute design has a large influence on the radial distribution of light
and heavy particles.
Without chute rotation (left figure), for a smooth chute (blue) the distribution is narrow, while
it becomes wider (red) for a rough chute and even wider still when a rough dividing wall is
inserted in the chute (green). This correlates with the increase in granular temperature at the
chute exit for the smooth and rough chute observed in Figure 6.14. Surprisingly, it does not
correlate with the relatively smaller granular temperature observed for the chute with a rough
dividing wall. We attribute this to the lower average velocity of the particles at the chute exit,
leading to a longer falling time and therefore a relatively stronger influence of the granular
130 Chapter 6
0.04 0.04
smooth chute
rough chute
0.03 insertion 0.03
θ (m2/s2)
0.02 0.02
0.01 0.01
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
1.1 1.1
1 1
α (−)
0.9 0.9
0.8 0.8
0.7 0.7
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Chute length(m)
Figure 6.14: Influence of roughness of chute and insertion of wall in the rough chute on
density segregation. Top row: averaged granular temperature along the length of the chute
for a rotation rate of 0 rpm (left) and 32 rpm (right). The symbols are for high density particles
and the lines for low density particles. Bottom row: ratio of fluctuating kinetic energies for a
rotation rate of 0 rpm (left) and 32 rpm (right).
10 10
smooth chute
8 rough chute 8
insertion
6 6
P(r)
4 4
2 2
0 0
1 1.5 2 1 1.5 2
Radial distance (m)
Figure 6.15: Radial distribution of low and high density particle on the burden surface plane
from the axis of rotation of chute for different chute geometries (see legend) at 0 rpm (left)
and 32 rpm (right). The symbols are for high density particles and the lines for low density
particles.
temperature at the chute exit. At the same time, the amount of segregation between heavy
and light particles is also increasing strongly when changing from a smooth chute (blue lines
vs blue symbols) to a rough chute (red) and rough chute with rough insert (green). This is in
agreement with the observed increase in segregation in Figure 6.13.
When the chute is rotating (right figure) all distributions move to a larger average distance
and get wider. It is interesting to note that under chute rotation, all peak positions get closer
together. As a result, the amount of segregation in the radial distribution is actually decreasing
with increasing chute rotation for the rough chute and chute with rough dividing wall.
6.8 Conclusions
Using a validated DEM model, we have investigated the influence of different process param-
eters on various quantities which are determining the rate of segregation and radial burden
distribution in a binary (density) mixture of granular particles flowing down a rotating chute.
We have found that segregation in the chute is enhanced by a higher granular temperature
and/or a higher density ratio. The granular temperature is increased when using a steeper in-
clination angle of the chute, by using a lower particle feeding mass rate, or by increasing the
chute roughness. We also found that chute rotation tends to increase the granular temperature
in the chute, and therefore the segregation in the chute. However, this segregation effect is
partly lost when considering the peak position of the radial burden distribution, which tend
to come closer together under the influence of chute rotation. We found that the width of
the radial burden distribution always increases with chute rotation. In view of developments
of continuum models based on kinetic theory of granular flows, we also investigated the
equipartition of fluctuation kinetic energy over the two different density phases. We found
that a perfect equipartition is generally not happening. Rather, the ratio α is very sensitive of
particle feeding mass rate, density ratio, and chute rotation. Finally, we found that the chute
design has a major influence on the segregation, both inside the chute and in the radial burden
distribution. Chute rotation can partly compensate this segregation effect in the radial burden
distribution.
Chapter
7
General conclusions and
recommendations
7.1 General conclusions

The objective of this thesis was to validate and modify a DEM simulation model tool to study
the dynamics of monodisperse and bidisperse granular flows down inclined rotating chutes.
We started our work, in chapter 2, with building an experimental setup and using this to
investigate monodisperse (3 mm glass bead) granular flows in rotating smooth chutes. The
importance of chute rotation can be estimated first through the Rossby number. For Rossby
numbers of the order of unity or smaller, the Coriolis forces start to dominate and particles
will move increasingly fast to the side wall. We found that the streamwise average surface
particle velocity is slightly reduced (by a few percent) in the upper section of the chute due
to compaction as a result of the Coriolis force, but strongly increases in the lower section
due to the increasing importance of the centrifugal forces. The particle bed height becomes
a two-dimensional function of the position inside the chute, with a strong increase in bed
height along the right sidewall due to the Coriolis forces. Both the bed height and width-wise
velocity show significant variations in the width direction due to the Coriolis forces.
Subsequently we did a comparative study, in chapter 3, of DEM simulations and experimental
results. We found that the precise inlet conditions are very important for the flow behaviour in
the initial part of the chute, but have no influence on the flow behaviour after a relatively short
distance from the inlet (in our case about 20 cm). We also found that the effect of air on the
flow behavior of the particles is relatively small and can be captured effectively and efficiently
by tuning the friction coefficient. For this type of chute flow, the most important particle
contact parameters are the friction coefficient and, to a lesser extent, the rotational friction
coefficient. The precise values of the normal and tangential coefficients of restitution have
133
134 Chapter 7
no discernible effect on the particle flow. We showed that the simulation model is capable
of quantitatively predicting the experimental results for streamwise and spanwise velocity of
particles at the surface and particle bed height in the chute. We highlighted how the simulation
model can subsequently be used to predict properties that are not readily accessible to optical
experiments such as solid volume fraction and shear flow profiles. We concluded that the
DEM simulations can be used with confidence to obtain more detailed insights in complex
flows such as mixtures of particles of different size or different density.
Disadvantage of the experiments described in chapter 3 is that they are very labour inten-
sive, especially with regard to the bed height measurements, where a full experiment needs
to be performed for every height measurement. In chapter 4 we investigated the ability to
apply three-dimensional particle tracking velocimetry (3D-PTV) to study bed height and sur-
face particle velocities of granular flow in a rotating chute. We cross-validated the 3D-PTV
technique with respect to these measurements against independent results obtained using an
electronic ultrasonic height sensor and particle image velocimetry (PIV). All experimental
results were found to be in good mutual agreement. The 3D-PTV technique has several ad-
vantages over other existing techniques. The advantages relative to the use of an electronic
height sensor for the measurement of the bed height are, first, a better spatial resolution of
the bed height profile and, second, a great reduction of the number of necessary experiments
(unless a large number of height sensors is used). The advantage of 3D-PTV relative to the
use of particle image velocimetry (PIV) is the measurement of the third (depthwise) compo-
nent of the surface particle velocity. An overall advantage of 3D-PTV is that it can provide
the measurements of particle bed height and surface particle velocity field within one and
the same experiment. We used the PTV experimental results to further validate our discrete
element model.
Next, in chapter 5 we used the validated DEM model to investigate the influence of base
roughness, in combination with chute rotation, on the behaviour of monodisperse and bidis-
perse (in density) granular flows through semi-cylindrical chutes. We found that the base
roughness has a strong influence on the bed height, average particle velocity, and granular
temperature of a monodisperse granular stream. The slip velocity with the chute wall is much
reduced for larger roughnesses, reaching an essentially zero value when the roughness is
twice the particle diameter. Rotation of the chute leads, besides the obvious sideways motion
of the particle stream, to an increase in bed height, an increase in average stream velocity,
a widening of the streamwise velocity profile, and an increase in granular temperature. The
base roughness also has a strong influence on the segregation rate in a binary density mixture.
For a smooth (but frictional) chute, the extent of segregation is essentially zero, it increases
strongly with increasing roughness, and finally saturates when the roughness is beyond one
particle diameter. Rotation of the chute tends to increase the segregation rate, but the in-
General conclusions and recommendations 135
crease is relatively small. An analysis of the ratio of fluctuating kinetic energies of the light
and heavy phase shows that an equipartition of energy almost holds. The ratio varies between
0.6 and 0.8, meaning that on average a larger fraction of the fluctuation kinetic energy is re-
siding in the dense phase. These observations will be important for future development and
validation of continuum methods applying to granular flows in chutes.
In chapter 6 we investigated the influence of various other process parameters on quantities
which are determining the rate of segregation in a binary (density) mixture of granular par-
ticles flowing down a rotating chute. We found, as expected, that the segregation rate in the
chute is enhanced by a higher granular temperature and/or a higher density ratio. The gran-
ular temperature is increased when using a steeper inclination angle of the chute, by using a
lower feeding mass rate, or by increasing the chute roughness. We found that chute rotation
tends to increase the granular temperature in the chute, and therefore the segregation in the
chute. However, this segregation effect is partly lost when considering the peak position of the
radial burden distribution, which tend to come closer together under the influence of chute
rotation. Finally, we found that the chute design has a major influence on the segregation,
both in the chute and in the radial burden distribution. Chute rotation can partly compensate
this segregation effect.
We also investigated, both experimentally and in DEM simulations, a binary mixture of small
(3 mm) and large (4.5 mm) particles of the same density, flowing through a smooth rectangu-
lar and semi-cylindrical chute (not shown in this thesis). Again the DEM results matched the
experiments well. We observed some segregation induced by the difference in size, but it was
not substantial. We concluded that segregation is negligible for a particle size ratio of 1.5,
at least within a 1 m long smooth chute. This is consistent with our observations on density
segregation in a smooth chute in chapter 5. All the binary size experiments and simulations
are world-wide unique and is a starting point for further research.
7.2 Recommendations
The experimental techniques and simulation model used in this thesis have proven successful
in validating of monodisperse flows down inclined rotating chutes. There are many possibil-
ities for further research on the topic of this thesis, and there are several possibilities to take
advantage of the developed experimental and simulation tools. We come to the following
recommendations based on priorities.
Different experimental and simulation approaches

1. The DEM model has been properly validated for monodisperse flows through inclined
rotating chutes for different chute shapes, different angles and different flow rates. Now
136 Chapter 7
it is much easier (and cheaper) to play with the DEM model to study different chute
geometries. This will give a good idea about how granular flows behave in different
chute geometries before actually performing the experiments. The same argument also
holds for binary size and density flows.
2. Before performing experiments on the rotating table, the above experiments should be
performed for a non-rotating chute with the PIV camera visualizing the flow through a
side wall of the chute. With this arrangement, segregation can be investigated and shear
flow profiles can be obtained, making use of the fact that friction with a rough bottom
wall is much larger than friction with a smooth side wall. This type of experiment is
difficult to perform on the rotating table because of the limited diameter of the table.
3. Binary size experiments: We performed binary size experiments with 1.5 size ratio
for different angles and flow rates in the rectangular and semi-cylindrical chute. We
used PTV technique for measurement of surface particle velocity and bed height in
the chute. We used colourless glass particles of size 4.5 mm large particle and 3 mm
small size particles and 3 mm blue particles as tracer particles in the experiments.
With this arrangement of experiments it turned out to be difficult to find segregation in
the experiments. We suggest to perform PTV experiments with two different colored
particles for different sizes. Also a granular mixture with higher difference in size
ratio of the particles, as well as rough base chutes, can be used to get considerable
segregation. The rough base can be achieved by glueing the particles having the same
properties but different diameters than the flowing particles on the flat plate and place
this plate in the rectangular chute.
4. In the DEM simulations, the initial arrangement of binary density mixtures is imple-
mented on the basis of a bcc crystal arrangement. For a binary size mixture a new
arrangement should be made in order to start with a random mixture of any size ratio
of particles at the inlet of chute.
5. Binary density experiments: The above experiments should be performed in the same
ways for different color of particles for different density. Different density ratios and
different chute roughnesses should be studied in PTV experiments.
6. The PTV technique used in the experiments is perfectly applicable to study the flow be-
haviour of granular materials through different chute geometries. We therefore suggest
to perform experiments for chutes different from the straight rectangular and semi-
cylindrical shapes studied here.
General conclusions and recommendations 137
7. The PTV setup calibration: The calibration method of PTV setup should be improved.
Instead of using a number of plates for different height during calibration of the PTV
setup, use of a single plate with more accurate moving arrangement should be used.
This will prevent errors due to wrong placement of the calibration plate.
8. The illumination in our PTV experiments was uneven and therefore we did not find
tracks in certain regions of the chute. A wider light source and a translucent paper
could be used to evenly lighten the entire span of the chute.
9. Granular temperature: Experiments should be performed to measure the granular tem-

perature for bi-disperse (in terms of size and density) granular flows in inclined rotating
chutes using an extension of PIV. Dijkhuizen et al. (2007) describe a method for such
a PIV extension. It should be noted that a higher resolution is necessary for these
measurements, compared to the resolution used for PIV in this work.
10. Experiments should be performed to also quantify the trajectories of particles falling
from the end of the chute onto the burden surface, for both rotating and non-rotating
chutes.
11. In the real blast furnace the coke, pellets and iron-ore particles have different shapes,
while currently we are using spherical particles in the simulations. The DEM model
should be extended to include non-spherical particles, for example by clumping to-
gether spherical particles.
Appendix
A
Appendix: Single particle
experiments
In this appendix, the explanation and results of the measurements for the coefficient of resti-
tution and the coefficient of friction are reported. Since the correct values of the coefficient
of restitution and coefficient of friction are highly important in the Discrete Element Model,
separate experiments are executed in order to obtain these values.
A.1 Coefficient of restitution

The coefficient of (normal) restitution represents the ratio of the velocity of two colliding
objects before and after the impact with each other. As particles collide many times with
other particles or with the wall of the chute, an accurate value for the coefficient of restitution
is necessary. The coefficient of restitution is dependent on the kind of material of both objects
and the impact geometry. In the case of the the particle is made of glass. The wall of the chute,
both for the rectangular and for the semi-cylindrical chute, is made of plexiglass. Thus, every
unique combination of these materials needs to be examined, with exception of plexiglass-
plexiglass as a collision of the wall with another wall is not possible.
The coefficient of restitution can be obtained by dropping a particle (material 1) on a plate
(material 2) and capture images of the particle before and after the bounces on the plate with
a high speed camera. Knowing the frame rate of the camera and the distance between two
frames gives the velocity of the particle. By dividing the velocity of the particle after the
collision with the velocity of the particle before the collision the coefficient of restitution
can be obtained. Figure A.1 shows the schematic representation of the experimental set-up
for the single particle experiment. The glass particle of diameter 3 mm and density of 2550
kg/m3 is used and two different plates are used (glass plate, and plexiglass plate) which
having dimensions of (W * D * H = 8*8*2 cm3 ). The frame rate at which images captured
139
140 Chapter A
Figure A.1: Schematic representation of the experimental set-up of the single particle exper-
iment (Schilde et al., 2011).
were 4000. For each combination of materials, at least 15 experiments were conducted to
reduce the experimental error. The final values for the reported coefficients of restitution
are the average of all individual experiments. Table A.1 shows the obtained coefficients of
restitution.
A.2 Coefficient of friction

As well as the coefficient of restitution, the coefficient of friction can be determined via
single particle experiments. Kharaz et al. (2001) describes a method to obtain the coefficient
of friction, based on both the coefficient of normal restitution and the coefficient of tangential
restitution. In the method for the coefficient of friction, not only experiments have to be
executed with a horizontal plate (which results in the coefficient of normal restitution), but
also with a tilted plate. From the experiments with a tilted plate, coefficients of tangential
restitution can be obtained dependent on the angle. The experiments were performed at
different inclination of plate from 0 degree to 80 degree with the interval of 10 degree. Kharaz
et al. (2001) found a relation between the tangential restitution coefficient et , the normal
restitution coefficient en , and the angle θ of the bottom plate. When plotting the tangential
restitution coefficient et , which is a function of the angle (θ ), against (1+en )cotθ , the slope of
the straight line through the points for the different angles represents the friction coefficient
Appendix: Single particle experiments 141
Figure A.2: Plots of et vs. (1+en )cotθ for glass-glass. The slope represents the friction
coefficient μ
Table A.1: Coefficients of (normal) restitution for all material combinations.
Material 1 and Material 2 Coefficient of restitution

Glass and Glass 0.9827 +/− 0.0017
Glass and Plexiglass 0.9457 +/− 0.0040
Table A.2: Coefficients of friction for material combinations.
Material 1 and Material 2 Coefficient of restitution

Glass and Glass 0.119
Glass and Plexiglass 0.097
μ. In this manner, the coefficients of friction for all combinations are computed. In Figure
A.2 the plots of et vs. (1+en )cotθ can be seen for two material combinations. The friction
coefficient, which follow from the plots, are given in Table A.2 .
Figure A.3 shows the dependency of restitution coefficient on the initial impact velocity of
particle while releasing on the plate. We found the restitution coefficient is within the error
bar which equal to 0.97 except at low velocity (top figure). we found similar observation
for effect of inclination of plate on the restitution coefficient and it is independent (bottom
figure).
142 Chapter 7
Figure A.3: Restitution coefficient dependency on the impact velocity (top figure) and incli-
nation angle ( bottom figure ) for glass particle and glass plate.
References
Akashi, M., Mio, H., Shimosaka, A., Shirakawa, Y., Hidaka, J., Nomura, S., 2008. Estimation
of bulk density distribution in particle charging process using discrete element method
considering particle shape. ISIJ international 48 (11), 1500–1506.
Alizadeh, E., Bertrand, F., Chaouki, J., 2014. Comparison of dem results and lagrangian
experimental data for the flow and mixing of granules in a rotating drum. AIChE Journal
60 (1), 60–75.
Alizadeh, E., Dubé, O., Bertrand, F., Chaouki, J., 2013. Characterization of mixing and size
segregation in a rotating drum by a particle tracking method. AIChE Journal 59 (6), 1894–
1905.
Ancey, C., 2001. Dry granular flows down an inclined channel: Experimental investigations
on the frictional-collisional regime. Physical Review E 65 (1), 011304.
Ancey, C., Coussot, P., Evesque, P., 1996. Examination of the possibility of a fluid-mechanics
treatment of dense granular flows. Mechanics of Cohesive-frictional Materials 1 (4), 385–
403.
Augenstein, D. A., Hogg, R., 1974. Friction factors for powder flow. Powder Technology
10 (1), 43–49.
Augenstein, D. A., Hogg, R., 1978. An experimental study of the flow of dry powders over
inclined surfaces. Powder Technology 19 (2), 205–215.
Baran, O., Ertaş, D., Halsey, T. C., Grest, G. S., Lechman, J. B., 2006. Velocity correlations
in dense gravity-driven granular chute flow. Physical Review E 74 (5), 051302.
Barbolini, M., Biancardi, A., Natale, L., Pagliardi, M., 2005. A low cost system for the
estimation of concentration and velocity profiles in rapid dry granular flows. Cold Regions
Science and Technology 43 (1), 49–61.
143
144
Beer, F. P., Johnson, E. R., 1973. Mechanics for Engineer, Static and dyanamics. New York,
MacGraw-Hill.
Beetstra, R., Van der Hoef, M. A., Kuipers, J. A. M., 2007. Drag force of intermediate
reynolds number flow past mono-and bidisperse arrays of spheres. AIChE Journal 53 (2),
489–501.
Bennett, S. J., Best, J. L., 1995. Particle size and velocity discrimination in a sediment-laden
turbulent flow using phase doppler anemometry. Journal of Fluids Engineering 117 (3),
505–511.
Bhattacharya, T., McCarthy, J., 2014. Chute flow as a means of segregation characterization.
Powder Technology 256, 126–139.
Boateng, A. A., Barr, P. V., 1997. Granular flow behaviour in the transverse plane of a par-
tially filled rotating cylinder. Journal of Fluid Mechanics 330, 233–249.
Bokkers, G. A., van Sint Annaland, M., Kuipers, J. A. M., 2004. Mixing and segregation in a
bidisperse gas–solid fluidised bed: a numerical and experimental study. Powder Technol-
ogy 140 (3), 176–186.
Bonamy, D., Daviaud, F., Laurent, L., 2002. Experimental study of granular surface flows
via a fast camera: a continuous description. Physics of Fluids (1994-present) 14 (5), 1666–
1673.
Börzsönyi, T., Ecke, R. E., 2006. Rapid granular flows on a rough incline: Phase diagram,
gas transition, and effects of air drag. Physical Review E 74 (6), 061301.
Brennen, C. E., Sieck, K., Paslaski, J., 1983. Hydraulic jumps in granular material flow.
Powder Technology 35 (1), 31–37.
Brown, R. L., 1939. The fundamental principles of segregation. The Institute of Fuel 13,
15–19.
Buist, K. A., Gaag, A. C., Deen, N. G., Kuipers, J. A. M., 2014. Improved magnetic particle
tracking technique in dense gas fluidized beds. AIChE Journal.
Campbell, C. S., Brennen, C. E., 1985a. Chute flows of granular material: some computer
simulations. Journal of applied mechanics 52 (1), 172–178.
Campbell, C. S., Brennen, C. E., 1985b. Computer simulation of granular shear flows. Journal
of Fluid Mechanics 151, 167–188.
145
Capart, H., Young, D. L., Zech, Y., 2002. Voronoï imaging methods for the measurement of
granular flows. Experiments in Fluids 32 (1), 121–135.
Chang, T. P., Wilcox, N. A., Tatterson, G. B., 1984. Application of image processing to the
analysis of three-dimensional flow fields. Optical Engineering 23 (3), 283–0.
Cheng, X., Lechman, J. B., Fernandez-Barbero, A., Grest, G. S., Jaeger, H. M., Karczmar,
G. S., Möbius, M. E., Nagel, S. R., 2006. Three-dimensional shear in granular flow. Phys-
ical review letters 96 (3), 038001.
Chevoir, F., Prochnow, M., Jenkins, J. T., Mills, P., 2001. Dense granular flows down an
inclined plane. Powders and grains 2001, 373.
Chou, H.-T., Lee, C.-F., 2009. Cross-sectional and axial flow characteristics of dry granular
material in rotating drums. Granular Matter 11 (1), 13–32.
Cleary, P. W., 2010. Dem prediction of industrial and geophysical particle flows. Particuology
8 (2), 106–118.
Cundall, P. A., Strack, O. D. L., 1979. A discrete numerical model for granular assemblies.
Geotechnique 29 (1), 47–65.
Daerr, A., Douady, S., 1999. Two types of avalanche behaviour in granular media. Nature
399 (6733), 241–243.
Dave, R. N., Rosato, A., Fischer, I. S., 1999. Non-intrusive particle tracking system for partic-
ulate flows and vibrated granular beds. Particulate science and technology 17 (1-2), 125–
139.
Del Castello, L., 2010. Table-top rotating turbulence: an experimental insight through particle
tracking. Ph.D. thesis, Eindhoven University of Technology, Eindhoven, The Netherlands.
Del Castello, L., Clercx, H. J. H., 2013. Geometrical statistics of the vorticity vector and the
strain rate tensor in rotating turbulence. Journal of Turbulence 14 (10), 19–36.
Delannay, R., Louge, M., Richard, P., Taberlet, N., Valance, A., 2007. Towards a theoretical
picture of dense granular flows down inclines. Nature Materials 6 (2), 99–108.
Deng, R., Wang, C.-H., 2003. Particle image velocimetry study on the pattern formation in a
vertically vibrated granular bed. Physics of Fluids (1994-present) 15 (12), 3718–3729.
Dent, J. D., Burrell, K. J., Schmidt, D. S., Louge, M. Y., Adams, E. E., Jazbutis, T. G.,
1998. Density, velocity and friction measurements in a dry-snow avalanche. Annals of
Glaciology 26, 247–252.
146
Di Maio, F. P., Di Renzo, A., Trevisan, D., 2009. Comparison of heat transfer models in dem-
cfd simulations of fluidized beds with an immersed probe. Powder Technology 193 (3),
257–265.
Dijkhuizen, W., Bokkers, G. A., Deen, N. G., Annaland, M. v., Kuipers, J. A. M., 2007.
Extension of piv for measuring granular temperature field in dense fluidized beds. AIChE
Journal 53 (1), 108–118.
Ding, Y. L., Seville, J. P. K., Forster, R., Parker, D. J., 2001. Solids motion in rolling mode
rotating drums operated at low to medium rotational speeds. Chemical Engineering Science
56 (5), 1769–1780.
Dippel, S., Wolf, D. E., 1999. Molecular dynamics simulations of granular chute flow. Com-
puter physics communications 121, 284–289.
Dolgunin, V. N., Kudy, A. N., Ukolov, A. A., 1998. Development of the model of segregation
of particles undergoing granular flow down an inclined chute. Powder technology 96 (3),
211–218.
Dolgunin, V. N., Ukolov, A. A., 1995. Segregation modeling of particle rapid gravity flow.
Drake, T. G., 1990. Structural features in granular flows. Journal of Geophysical Research:
Solid Earth (1978–2012) 95 (B6), 8681–8696.
Forterre, Y., Pouliquen, O., 2001. Longitudinal vortices in granular flows. Physical review
letters 86 (26), 5886.
Fuchs, R., Weinhart, T., Meyer, J., Zhuang, H., Staedler, T., Jiang, X., Luding, S., 2014.
Rolling, sliding and torsion of micron-sized silica particles: experimental, numerical and
theoretical analysis. Granular Matter 16 (3), 281–297.
Gidaspow, D., 1994. Multiphase flow and fluidization: continuum and kinetic theory descrip-
tions. Academic press.
Godlieb, W., 2010. High pressure fluidization. Ph.D. thesis, Eindhoven University of Tech-
nology, Eindhoven, The Netherlands.
Goldschmidt, M. J. V., 2001. Hydrodynamic modelling of fluidised bed spray granulation.

Ph.D. thesis, University of Twente, Enschede, The Netherlands.
Goldschmidt, M. J. V., Kuipers, J. A. M., Van Swaaij, W. P. M., 2001. Hydrodynamic mod-
elling of dense gas-fluidised beds using the kinetic theory of granular flow: effect of coef-
ficient of restitution on bed dynamics. Chemical Engineering Science 56 (2), 571–578.
147
Goujon, C., Thomas, N., Dalloz-Dubrujeaud, B., 2003. Monodisperse dry granular flows on
inclined planes: Role of roughness. The European Physical Journal E: Soft Matter and
Biological Physics 11 (2), 147–157.
Gray, J. M. N. T., Tai, Y. C., Noelle, S., 2003. Shock waves, dead zones and particle-free
regions in rapid granular free-surface flows. Journal of Fluid Mechanics 491, 161–181.
Guler, M., Edil, T. B., Bosscher, P. J., 1999. Measurement of particle movement in granular
soils using image analysis. Journal of Computing in Civil Engineering 13 (2), 116–122.
Hákonardóttir, K. M., Hogg, A. J., 2005. Oblique shocks in rapid granular flows. Physics of
Fluids (1994-present) 17 (7), 077101.
Hanes, D. M., Walton, O. R., 2000. Simulations and physical measurements of glass spheres
flowing down a bumpy incline. Powder technology 109 (1), 133–144.
Hill, K. M., Caprihan, A., Kakalios, J., 1997. Bulk segregation in rotated granular material
measured by magnetic resonance imaging. Physical Review Letters 78 (1), 50.
Ho, C. K., Wu, S. M., Zhu, H. P., Yu, A. B., Tsai, S. T., 2009. Experimental and numerical
investigations of gouge formation related to blast furnace burden distribution. Minerals
Engineering 22 (11), 986–994.
Holyoake, A. J., 2011. Rapid granular flows in an inclined chute. Ph.D. thesis, University of
Cambridge, Cambridge, England, United Kingdom.
Holyoake, A. J., McElwaine, J. N., 2012. High-speed granular chute flows. Journal of Fluid
Mechanics 710, 35–71.
Hoomans, B., Kuipers, J., Briels, W., Van Swaaij, W., 1996. Discrete particle simulation of
bubble and slug formation in a two-dimensional gas-fluidised bed: a hard-sphere approach.
Chemical Engineering Science 51 (1), 99–118.
Hoomans, B. P. B., Kuipers, J. A. M., Mohd Salleh, M. A., Stein, M., Seville, J. P. K.,
2001. Experimental validation of granular dynamics simulations of gas-fluidised beds with
homogenous in-flow conditions using positron emission particle tracking. Powder Tech-
nology 116 (2), 166–177.
Ishida, M., Shirai, T., 1979. Velocity distributions in the flow of solid particles in an inclined
open channel. J. Chem. Eng. Jpn. 12.
Jaeger, H. M., Nagel, S. R., Behringer, R. P., 1996. Granular solids, liquids, and gases. Re-
views of Modern Physics 68 (4), 1259–1273.
148
Jaeger, H. M., Nagel, S. R., et al., 1992. Physics of the granular state. Science 255 (5051),
1523–1531.
Jain, N., Ottino, J. M., Lueptow, R. M., 2002. An experimental study of the flowing granular
layer in a rotating tumbler. Physics of Fluids (1994-present) 14 (2), 572–582.
Jasti, V., Higgs III, C. F., 2008. Experimental study of granular flows in a rough annular shear
cell. Physical Review E 78 (4), 041306.
Jenkins, J. T., 1992. Boundary conditions for rapid granular flow: flat, frictional walls. Journal
of applied mechanics 59 (1), 120–127.
Johnson, P. C., Nott, P., Jackson, R., 1990. Frictional–collisional equations of motion for
participate flows and their application to chutes. Journal of Fluid Mechanics 210, 501–
535.
Kajiwara, Y., Jimbo, T., JOKO, T., AMINAGA, Y.-i., INADA, T., 1984. Investigation of bell-
less charging based on full scale model experiments. Transactions of the Iron and Steel
Institute of Japan 24 (10), 799–807.
Kaneko, Y., Shiojima, T., Horio, M., 1999. Dem simulation of fluidized beds for gas-phase
olefin polymerization. Chemical Engineering Science 54 (24), 5809–5821.
Kano, J., Kasai, E., Saito, F., Kawaguchi, T., 2005. Numerical simulation model for granula-
tion kinetics of iron ores. ISIJ international 45 (4), 500–505.
Kawaguchi, T., 2010. Mri measurement of granular flows and fluid-particle flows. Advanced
Keane, R. D., Adrian, R. J., 1991. Optimization of particle image velocimeters: Ii. multiple
pulsed systems. Measurement science and technology 2 (10), 963.
Khakhar, D. V., McCarthy, J. J., Ottino, J. M., 1999. Mixing and segregation of granular
materials in chute flows. Chaos: An Interdisciplinary Journal of Nonlinear Science 9 (3),
594–610.
Kharaz, A. H., Gorham, D. A., Salman, A. D., 2001. An experimental study of the elastic
rebound of spheres. Powder Technology 120 (3), 281–291.
Kostek, R., Landowski, B., 2011. A simulation study of mixing granular materials. Journal
of Polish CIMAC 6, 57–64.
Kou, M., Wu, S., Du, K., Shen, W., Sun, J., Zhang, Z., 2013. Dem simulation of burden
distribution in the upper part of corex shaft furnace. ISIJ international 53 (6), 1002–1009.
149
Krcek, B., Bocka, V., Broz, L., Mrazek, J., 1977. Mathematical modelling of the motion of
burden materials in a blast furnace. Hutnicke Listy 32 (7), 466–470.
Kumaran, V., Bharathraj, S., 2013. The effect of base roughness on the development of
a dense granular flow down an inclined plane. Physics of Fluids (1994-present) 25 (7),
070604.
Kundu, P. K., Cohen, I. M., 2008. Fluid Mechanics. Elsevier Academic Press.
Laverman, J. A., 2010. On the hydrodynamics in gas phase polymerization reactors. Ph.D.
thesis, Eindhoven University of Technology, Eindhoven, The Netherlands.
Laverman, J. A., Roghair, I., Annaland, M. v. S., Kuipers, H., 2008. Investigation into the
hydrodynamics of gas–solid fluidized beds using particle image velocimetry coupled with
digital image analysis. The Canadian Journal of Chemical Engineering 86 (3), 523–535.
Li, Q., Feng, M., Zou, Z., 2013. Validation and calibration approach for discrete element
simulation of burden charging in pre-reduction shaft furnace of corex process. ISIJ inter-
national 53 (8), 1365–1371.
Liao, C.-C., Hsiau, S.-S., 2009. Influence of interstitial fluid viscosity on transport phe-
nomenon in sheared granular materials. Chemical Engineering Science 64 (11), 2562–
2569.
Link, J. M., Zeilstra, C., Deen, N. G., Kuipers, J. A. M., 2004. Validation of a discrete
particle model in a 2d spout-fluid bed using non-intrusive optical measuring techniques.
The Canadian Journal of Chemical Engineering 82 (1), 30–36.
Louge, M. Y., Steiner, R., Keast, S. C., Decker, R., Dent, J., Schneebeli, M., 1997. Applica-
tion of capacitance instrumentation to the measurement of density and velocity of flowing
snow. Cold regions science and Technology 25 (1), 47–63.
Luding, S., 2008. Cohesive, frictional powders: contact models for tension. Granular matter
10 (4), 235–246.
Lueptow, R. M., Akonur, A., Shinbrot, T., 2000. Piv for granular flows. Experiments in Fluids
28 (2), 183–186.
Maas, H. G., 1991. Digital photogrammetry for determination of tracer particle coordinates in
turbulent flow research. Photogrammetric engineering and remote sensing 57 (12), 1593–
1598.
150
Maas, H. G., 1995. New developments in multimedia photogrammetry. Optical 3D measure-

ment techniques III.
Maas, H. G., Gruen, A., Papantoniou, D. A., 1993. Particle tracking velocimetry in three-
dimensional flows. Part I : Photogrammetric determination of particle coordinates. Exper-
iments in Fluids 15 (2), 133–146.
Maas, H. G., Virant, M., Becker, J., Bösemann, W., Gatti, L., Henrichs, A., 2002. Photogram-
metric methods for measurements in fluid physics experiments in space. Acta Astronautica
50 (4), 225–231.
Malik, N. A., Dracos, T. H., Papantoniou, D. A., 1993. Particle tracking velocimetry in three-
dimensional flows. Part II : Particle tracking. Experiments in Fluids 15 (4-5), 279–294.
Mellmann, J., 2001. The transverse motion of solids in rotating cylinders forms of motion
and transition behavior. Powder Technology 118 (3), 251–270.
Michaels, J. N., 2003. Toward rational design of powder processes. Powder technology
138 (1), 1–6.
MiDia, G. D. R., 2004. On dense granular flows. Eur. Phys. J. E 14, 341–365.
Mio, H., Kadowaki, M., Matsuzaki, S., Kunitomo, K., 2012. Development of particle flow
simulator in charging process of blast furnace by discrete element method. Minerals Engi-
neering 33, 27–33.
Mio, H., Komatsuki, S., Akashi, M., Shimosaka, A., Shirakawa, Y., Hidaka, J., Kadowaki,
M., Matsuzaki, S., Kunitomo, K., 2008. Validation of particle size segregation of sintered
ore during flowing through laboratory-scale chute by discrete element method. ISIJ inter-
national 48 (12), 1696–1703.
M., Matsuzaki, S., Kunitomo, K., 2009. Effect of chute angle on charging behavior of
sintered ore particles at bell-less type charging system of blast furnace by discrete element
method. ISIJ international 49 (4), 479–486.
M., Yokoyama, H., Matsuzaki, S., Kunitomo, K., 2010. Analysis of traveling behavior of
nut coke particles in bell-type charging process of blast furnace by using discrete element
151
Mio, H., Yamamoto, K., Shimosaka, A., Shirakawa, Y., Hidaka, J., 2007. Modeling of solid
particle flow in blast furnace considering actual operation by large-scale discrete element
Müller, C. R., Holland, D. J., Sederman, A. J., Scott, S. A., Dennis, J. S., Gladden, L. F.,
2008. Granular temperature: comparison of magnetic resonance measurements with dis-
crete element model simulations. Powder Technology 184 (2), 241–253.
Müller, C. R., Scott, S. A., Holland, D. J., Clarke, B. C., Sederman, A. J., Dennis, J. S.,
Gladden, L. F., 2009. Validation of a discrete element model using magnetic resonance
measurements. Particuology 7 (4), 297–306.
Mullin, T., 2002. Mixing and de-mixing. Science 295 (5561), 1851–1851.
Nag, S., Koranne, V. M., 2009. Development of material trajectory simulation model for blast
furnace compact bell-less top. Ironmaking & Steelmaking 36 (5), 371–378.
Nakagawa, M., Altobelli, S. A., Caprihan, A., Fukushima, E., Jeong, E. K., 1993. Non-
invasive measurements of granular flows by magnetic resonance imaging. Experiments in
fluids 16 (1), 54–60.
Nedderman, R. M., 2005. Statics and kinematics of granular materials. Ph.D. thesis, Univer-
sity of Cambridge, Cambridge, England, United Kingdom.
Nguyen, V. D., Cogné, C., Guessasma, M., Bellenger, E., Fortin, J., 2009. Discrete model-
ing of granular flow with thermal transfer: application to the discharge of silos. Applied
thermal engineering 29 (8), 1846–1853.
Nieuwland, J. J., van Sint Annaland, M., Kuipers, J. A. M., Van Swaaij, W. P. M., 1996. Hy-
drodynamic modeling of gas/particle flows in riser reactors. AIChE Journal 42 (6), 1569–
1582.
Nouchu, T., Sato, M., Takeda, K., Ariyama, T., 2005. Effects of operation condition and cast-
ing strategy on drainage efficiency of the blast furnace hearth. ISIJ international 45 (10),
1515–1520.
Ottino, J. M., Khakhar, D. V., 2000. Mixing and segregation of granular materials. Annual
Review of Fluid Mechanics 32, 55–91.
Ottino, J. M., Khakhar, D. V., 2001. Fundamental research in heaping, mixing, and seg-
regation of granular materials: challenges and perspectives. Powder technology 121 (2),
117–122.
152
Ottino, J. M., Lueptow, R. M., 2008. On mixing and demixing. SCIENCE-NEW YORK
THEN WASHINGTON- 319 (5865), 912.
Parker, D. J., Dijkstra, A. E., Martin, T. W., Seville, J. P. K., 1997. Positron emission parti-
cle tracking studies of spherical particle motion in rotating drums. Chemical Engineering
Science 52 (13), 2011–2022.
Pohlman, N. A., Meier, S. W., Lueptow, R. M., Ottino, J. M., 2006. Surface velocity in three-
dimensional granular tumblers. Journal of Fluid Mechanics 560, 355–368.
Pouliquen, O., 1999. Scaling laws in granular flows down rough inclined planes. Physics of
Fluids (1994-present) 11 (3), 542–548.
Pouliquen, O., Forterre, Y., 2002. Friction law for dense granular flows: application to the
motion of a mass down a rough inclined plane. Journal of Fluid Mechanics 453, 133–151.
Pouliquen, O., Renaut, N., 1996. Onset of granular flows on an inclined rough surface: dila-
tancy effects. Journal de Physique II 6 (6), 923–935.
Prasad, A. K., 2000. Particle image velocimetry. Current Science - Bangalore 79 (1), 51–60.
Pudasaini, S. P., Hutter, K., 2007. Avalanche dynamics: dynamics of rapid flows of dense
granular avalanches. Springer.
Pudasaini, S. P., Hutter, K., Hsiau, S.-S., Tai, S.-C., Wang, Y., Katzenbach, R., 2007. Rapid
flow of dry granular materials down inclined chutes impinging on rigid walls. Physics of
Fluids (1994-present) 19 (5), 053302.
Racca, R. G., M, D. J., 1988. A method for automatic particle tracking in a three-dimensional
flow field. Experiments in Fluids 6 (1), 25–32.
Radhakrishnan, V. R., Maruthy Ram, K., 2001. Mathematical model for predictive control
of the bell-less top charging system of a blast furnace. Journal of Process Control 11 (5),
565–586.
Ristow, G. H., 1996. Dynamics of granular materials in a rotating drum. EPL (Europhysics
Letters) 34 (4), 263.
Roberts, A. W., 1969. An investigation of the gravity flow of noncohesive granular materials
through discharge chutes. Journal of Manufacturing Science and Engineering 91 (2), 373–
381.
Roberts, A. W., 2003. Chute performance and design for rapid flow conditions. Chemical
engineering & technology 26 (2), 163–170.
153
Roberts, A. W., Scott, O. J., 1981. Flow of bulk solids through transfer chutes of variable
geometry and profile. Bulk Solids Handling 1.
Rosenqvist, T., 2004. Principles of extractive metallurgy. Tapir Academic Press.
Sakaguchi, H., Ozaki, E., Igarashi, T., 1993. Plugging of the flow of granular materials during
the discharge from a silo. International Journal of Modern Physics B 7 (09n10), 1949–
1963.
Santana, D., Nauri, S., Acosta, A., García, N., Macías-Machín, A., 2005. Initial particle
velocity spatial distribution from 2d erupting bubbles in fluidized beds. Powder Technology
150 (1), 1–8.
Savage, S., Lun, C., 1988. Particle size segregation in inclined chute flow of dry cohesionless
granular solids. Journal of Fluid Mechanics 189, 311–335.
Savage, S. B., Hutter, K., 1991. The dynamics of avalanches of granular materials from
initiation to runout. part i: Analysis. Acta Mechanica 86 (1-4), 201–223.
Savage, S. B., McKeown, S., 1983. Shear stresses developed during rapid shear of concen-
trated suspensions of large spherical particles between concentric cylinders. Journal of
Fluid Mechanics 127, 453–472.
Sawley, M. L., Zaimi, S. A., Sert, D., 2011. Study of the charging of a blast furnace with a ro-
tating chute using the discrete element method. In: Proceedings of Particle Based Methods
II, Fundamentals and Applications.
Schilde, C., Beinert, S., Kwade, A., 2011. Comparison of the micromechanical aggregate
properties of nanostructured aggregates with the stress conditions during stirred media
milling. Chemical Engineering Science 66 (21), 4943–4952.
Sela, N., Goldhirsch, I., 1998. Hydrodynamic equations for rapid flows of smooth inelastic
spheres, to burnett order. Journal of Fluid Mechanics 361, 41–74.
Sheng, L.-T., 2012. Dynamics of granular avalanches: An experimental and numerical study.
Ph.D. thesis, National Central University, Taiwan.
Sherritt, R. G., Chaouki, J., Mehrotra, A. K., Behie, L. A., 2003. Axial dispersion in the
three-dimensional mixing of particles in a rotating drum reactor. Chemical Engineering
Science 58 (2), 401–415.
Shirsath, S. S., Padding, J. T., Clercx, H. J. H., Kuipers, J. A. M., 2012. Modelling of granular
flows through inclined rotating chutes using a discrete particle model.
154
Shirsath, S. S., Padding, J. T., Clercx, H. J. H., Kuipers, J. A. M., 2014a. Cross-validation of
3d particle tracking velocimetry for the study of granular ows down rotating chutes. CES
Journal.
Shirsath, S. S., Padding, J. T., Deen, N. G., Clercx, H. J. H., Kuipers, J. A. M., 2013. Exper-
imental study of monodisperse granular flow through an inclined rotating chute. Powder
technology 246, 235–246.
Shirsath, S. S., Padding, J. T., Peeters, T. W. J., Clercx, H. J. H., Kuipers, J. A. M., 2014b.
Numerical investigation of monodisperse granular flow through an inclined rotating chute.
AIChE Journal 60 (10), 3424–3441.
Silbert, L. E., Ertaş, D., Grest, G. S., Halsey, T. C., Levine, D., Plimpton, S. J., 2001. Granular
flow down an inclined plane: Bagnold scaling and rheology. Physical Review E 64 (5),
051302.
Standish, N., 1979. Principles in burdening and bell-less charging. Nimaroo Publishers, 110.
Stewart, R. L., Bridgwater, J., Zhou, Y. C., Yu, A. B., 2001. Simulated and measured flow of
granules in a bladed mixer?a detailed comparison. Chemical Engineering Science 56 (19),
5457–5471.
Sun, J., 2007. Multiscale modeling of segregation in granular flows. Ph.D. thesis, Iowa State
University, Iowa, United States.
Thornton, A., Weinhart, T., Luding, S., Bokhove, O., 2012a. Modeling of particle size seg-
regation: Calibration using the discrete particle method. International Journal of Modern
Physics C 23 (08).
Thornton, A. R., Weinhart, T., Luding, S., Bokhove, O., 2012b. Frictional dependence of
shallow-granular flows from discrete particle simulations.
Van Bokhoven, L. J. A., Clercx, H. J. H., Van Heijst, G. J. F., Trieling, R. R., 2009. Experi-
ments on rapidly rotating turbulent flows. Physics of Fluids (1994-present) 21 (9), 096601.
Van der Hoef, M. A., Ye, M., van Sint Annaland, M., Andrews, A. T., Sundaresan, S.,
Kuipers, J. A. M., 2006. Multiscale modeling of gas-fluidized beds. Advances in Chemical
Engineering 31, 65–149.
Van Heijst, G. J. F., Clercx, H. J. H., 2009. Laboratory modeling of geophysical vortices.
Annual Review of Fluid Mechanics 41, 143–164.
155
Veje, C. T., Howell, D. W., Behringer, R. P., 1999. Kinematics of a two-dimensional granular
couette experiment at the transition to shearing. Physical Review E 59 (1), 739.
Vreman, A. W., Al-Tarazi, M., Kuipers, J. A. M., van Sint Annaland, M., Bokhove, O.,
2007. Supercritical shallow granular flow through a contraction: experiment, theory and
simulation. Journal of Fluid Mechanics 578, 233–269.
Weinhart, T., Thornton, A. R., Luding, S., Bokhove, O., 2012. Closure relations for shallow
granular flows from particle simulations. Granular matter 14 (4), 531–552.
Westerweel, J., 1997. Fundamentals of digital particle image velocimetry. Measurement Sci-
ence and Technology 8 (12), 1379.
Wibowo, C., Ng, K. M., 2001. Operational issues in solids processing plants: systems view.
AIChE journal 47 (1), 107–125.
Wightman, C., Moakher, M., Muzzio, F. J., Walton, O., 1998. Simulation of flow and mixing
of particles in a rotating and rocking cylinder. AIChE Journal 44 (6), 1266–1276.
Williams, J. C., 1976. The segregation of particulate materials. a review. Powder technology
15 (2), 245–251.
Willneff, J., 2002. 3d particle tracking velocimetry based on image and object space infor-
mation. International archives of photogrammetry remote sensing and spatial information
science 34 (5), 601–606.
Willneff, J., 2003. A new spatio-temporal matching algorithm for 3d particle tracking ve-
locimetry. Ph.D. thesis, Swiss Federal Institute of Technology, Zurich, Switzerland.
Willneff, J., Grün, A., Grün, A., Grün, A., 2012. A new spatio-temporal matching algorithm
for 3d-particle tracking velocimetry. In: Proceedings of the 9th International Symposium
on Transport Phenomena and Dynamics of Rotating Machinery, Honolulu, Hawaii, USA,
February.
Wu, S., Kou, M., Xu, J., Guo, X., Du, K., Shen, W., Sun, J., 2013. Dem simulation of particle
size segregation behavior during charging into and discharging from a paul-wurth type
hopper. Chemical Engineering Science 99, 314–323.
Yang, S. C., 2006. Density effect on mixing and segregation processes in a vibrated binary
granular mixture. Powder Technology 164 (2), 65–74.
Yang, W.-L., Hsiau, S.-S., 2006. The effect of liquid viscosity on sheared granular flows.
Chemical engineering science 61 (18), 6085–6095.
156
Yu, Y., 2008. Study and develop on the model for bell-less charging of blast furnace. Master’s
thesis, Chongqing University, Chongqing, China.
Yu, Y., 2013. Experimental and discrete element simulation studies of bell-less charging of
blast furnace. Ph.D. thesis, Åbo Akademi University, Åbo, Finland.
Yu, Y., Saxén, H., 2010. Experimental and dem study of segregation of ternary size particles
in a blast furnace top bunker model. Chemical engineering science 65 (18), 5237–5250.
Yu, Y., Saxén, H., 2011a. Discrete element method simulation of properties of a 3d conical
hopper with mono-sized spheres. Advanced powder technology 22 (3), 324–331.
Yu, Y., Saxen, H., 2012. Flow of pellet and coke particles in and from a fixed chute. Industrial
& Engineering Chemistry Research 51 (21), 7383–7397.
Yu, Y., Saxén, H., 2013. Particle flow and behavior at bell-less charging of the blast furnace.
Steel Research International 84 (10), 1018–1033.
Yu, Y. W., Saxén, H., 2011b. Analysis of rapid flow of particles down and from an inclined
chute using small scale experiments and discrete element simulation. Ironmaking & Steel-
making 38 (6), 432–441.
Zaimi, S. A., Campos, T., Bennani, M., Lecacheux, B., Danloy, G., Pomeroy, D., Perez-
Chust, R., 2009. Blast furnace models development and application in arcelormittal group.
Revue de Métallurgie 106 (03), 105–111.
Zeilstra, C., 2007. Granular dynamics in vibrated beds. Ph.D. thesis, University of Twente,
Enschede, The Netherlands.
Zeilstra, C., Collignon, J. G., Van der Hoef, M. A., Deen, N. G., Kuipers, J. A. M., 2008. Ex-
perimental and numerical study of wall-induced granular convection. Powder Technology
184 (2), 166–176.
Zenit, R., Hunt, M. L., Brennen, C. E., 1997. Collisional particle pressure measurements in
solid–liquid flows. Journal of Fluid Mechanics 353, 261–283.
Zhang, J., Hu, Z., Ge, W., Zhang, Y., Li, T., Li, J., 2004. Application of the discrete approach
to the simulation of size segregation in granular chute flow. Industrial & engineering chem-
istry research 43 (18), 5521–5528.
Zhang, J., Qiu, J., Guo, H., Ren, S., Sun, H., Wang, G., Gao, Z., 2014. Simulation of parti-
cle flow in a bell-less type charging system of a blast furnace using the discrete element
method. Particuology.
157
Zheng, X. M., Hill, J. M., 1998. Molecular dynamics simulation of granular flows: Slip along
rough inclined planes. Computational mechanics 22 (2), 160–166.
Zhou, Y. C., Wright, B. D., Yang, R. Y., Xu, B. H., Yu, A. B., 1999. Rolling friction in
the dynamic simulation of sandpile formation. Physica A: Statistical Mechanics and its
Applications 269 (2), 536–553.
Zhou, Z., Zhu, H., Yu, A., Wright, B., Pinson, D., Zulli, P., 2005. Discrete particle simulation
of solid flow in a model blast furnace. ISIJ international 45 (12), 1828–1837.
Zhou, Z. Y., Zhu, H. P., Wright, B., Yu, A. B., Zulli, P., 2011. Gas–solid flow in an ironmaking
blast furnace-ii: Discrete particle simulation. Powder Technology 208 (1), 72–85.
Nomenclature
Roman symbols
A, B constants (−)
D diameter (m)
dt time step (s)
e restitution coefficient (−)
F force (N)
Fr (rotational) Froude number (−)
g gravitational acceleration (m/s2 )
I moment of inertia (kg.m2 )
k spring stiffness (N/m)
L length of chute (−)
m mass of particle (kg)
N number of particles (−)
NX, NY, NZ number of grid cells in x,y,z-direction (−)
n normal unit vector (−)
p gas pressure (Pa)
Q mass flow rate (kg/s)
159
160
Rp particle radius (m)
Ro Rossby number (−)
Rr rolling radius (m)
r position (m)
Sp momentum source term N/m3
T torque (Nm)
Tr rolling torque (Nm)
t tangential unit vector (−)
t time (s)
u gas velocity (m/s)
V local volume of the computational cells (m3 )
Va volume of particle (m3 )
v particle velocity (m/s)
Greek symbols
α ratio of fluctuating kinetic energy (−)
β interphase momentum transfer coefficient (kg/m3 .s)
δ overlap vector (m)
δ distribution function (−)
ε local porosity (−)
η damping coefficient (−)
θ total granular temperature (−)
λ gas-phase bulk viscosity (kg/m.s)
μ coefficient of friction (−)

161
μr coefficient of rolling friction (−)
μg gas phase shear viscosity (kg/m.s)
ρ density (kg/m3 )
τ gas-phase stress tensor (kg/m.s2 )
φ angle of inclination of the chute (degree)
ω rotational velocity of a particle (rad/s)
Ω rotation rate of chute (rad/s)
Subscripts
a, b particle indices
d heavy phase
g gas phase
l light phase
n normal
p particle
rel relative
t tangential
tot total
w wall
x, y, z in x, y, z-directions
Superscripts
c contact
d drag
p pressure
T transposed
162
Abbreviations
2D Two-dimensional
3D Three-dimensional
DEM Discrete element model
MPT Magnetic particle tracking
MRI Magnetic resonanace imaging
PIV Particle image velocimetry
PTV Particle tracking velocimetry
PEPT Positron emission particle tracking

Publications
Journal Publications
1. Shirsath, S.S., Padding, J.T., Deen, N.G., Clercx, H.J.H. and Kuipers, J.A.M. (2013).
Experimental study of monodisperse granular flow through an inclined rotating chute.
Powder Technology, 246, 235-246
2. Shirsath, S.S., Padding, J.T., Peeters, T.W.J., Clercx, H.J.H. and Kuipers, J.A.M. (2014).
Numerical investigation of monodisperse granular flow through an inclined rotating
chute. AIChE Journal, 60(10), 3424-3441.
3. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. Cross-validation of 3D
Particle Tracking Velocimetry with other technique for granular flows down rotating
chutes. Submitted to CES Journal.
4. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. Simulation study
of the effect of wall roughness on the dynamics of granular lows in rotating semi-
cylindrical chutes. Submitted to AIChE Journal.
Conference Publications
1. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. (2012). Modelling of
granular flows through inclined rotating chutes using a discrete particle model. Confer-
ence Paper at the Ninth International Conference on CFD in the Minerals and Process
Industries CSIRO, 10-12 December 2012, Melbourne, Australia.
2. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. (2014). Dynamics
of granular flows down rotating semi-cylindrical chutes Conference Paper at the 7th
World Congress for Particle Technology, 19-22 May 2014, Beijing, China.
163
164
Oral Presentations
1. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. Modelling of granular
flows through inclined rotating chutes using a discrete particle model. Oral presentation
at the Ninth International Conference on CFD in the Minerals and Process Industries
CSIRO, 10-12 December 2012, Melbourne, Australia.
2. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. Experimental and
computational investigation on the flow behavior of granular particles through an in-
clined rotating chute. Oral : presentation at the APS DFD Meeting, 24-26 November
2013, Pittsburgh, USA.
3. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. Dynamics of granular
flows down rotating semi-cylindrical chutes Oral : presentation at the WCPT7, 19-22
May 2014, Beijing, China.
4. Clercx, H.J.H., Shirsath, S.S., Padding, J.T. and Kuipers, J.A.M. Three-dimensional
particle tracking velocimetry applied to granular flows down rotating chutes. Oral :
presentation at the APS DFD Meeting 23-25 November 2014, San Francisco, CA,
USA.
5. Shirsath, S.S., Padding, J.T., Clercx, H.J.H. and Kuipers, J.A.M. Cross-validation of 3D
particle tracking velocimetry and separation of granular mixture flows down rotating
chutes Oral : presentation at the FOM Meeting, 20-21 January 2015, Veldhoven, The
Netherlands.
Acknowledgments
This dissertation would not have been possible without the guidance and help of some indi-
viduals who in one way or another contributed and extended their invaluable assistance in the
preparation and completion of this work. Although they are too numerous to mention, I must,
however, identify key individuals whose contribution were obvious.
First of all, I would like to thank prof. Hans Kuipers and prof. Herman Clercx for giving
me the opportunity to perform my PhD thesis research under their supervision. Without their
enthusiasm, encouragement, support and continuous optimism this thesis would hardly have
been completed. I am very grateful their advice on how to develop into a good researcher. I
express my warmest gratitude to my daily supervisor dr. Johan Padding. His guidance and
help in modelling has been a valuable input for this thesis.
I also want to thank to dr. Martin van der Hoef for helping me as a daily supervisor in the
initial phase of my PhD. I would also like to thank dr. Niels Deen for helping me by giving
idea on construction of setup and also for understanding PIV technique.
My special thanks go to my professors in ICT Mumbai, prof. A.V. Patwardhan and prof.
A.B. Pandit. I am grateful that they expressed their confidence in me and forwarded my
application in this group with a good recommendation for me. This all would not have been
possible without their help.
I would like to express my gratitude to Dutch Technology Foundation STW and our industrial
partner Tata Steel, IJmuiden. Also I would like to thank all the committee members for their
valuable time in reviewing this thesis and accepting to be board on the day of the PhD defense.
I would like to thank Prof. Onno Bokhove, Dr. Anthony Thornton and Deepak Tunuguntla
from Twente University for stimulating discussions during STW project meeting.
My gratitude goes to our secretaries José, Ada and Judith for their help in administration
matters, arranging all documentation related to conferences.
I would like to thank Yuk Man for helping me initially with visualization and postprocessing
tools. I also like to thank Martin Korevaar for helping me in understanding and coding matlab
165
166
script for the plotting the graphs.

My sincere regards to my M.Sc. students Merijn Buijs and Vinay Mahajan for their experi-
mental and simulation work.
I wish to express my thanks to our group technician Lee McAlpine, who played important
role in constructing the chute flow setup and transporting it to the Applied physics department
for experiments on the rotating table. I would like thanks Ad Holten from Physics department
for helping me in arranging the PTV camera facility on the rotating table and modifying the
ETH Zurich code for granular flows system.
I would like to thank all SMR group members and students for the pleasant time during coffee
breaks, group lectures, borrels and group outings. Special thanks go to Hamid, Yali, Lucia,
Paul, Yupeng, Nhi and Younas for nice non-technical discussions.
I would like to take this opportunity to thank all my dear friends in the Netherlands from
India for their cooperation, enjoyment and beautiful memories. My special thanks to Sandip-
Priyanka, Gajanan-Vrushali, Amit-Dipali, Amit-Kiranmayee, Kamlesh-Swati, Deepak, Vinayak,
Vikrant, Hitesh, Monali, Yogesh, Narendra, Bhaskar, Srivathsa, Saurish, Shauvik, Rohit,
Sathish, Krushna, Faiyas, Amol, Jagdish and Pranav.
Also, I thank my friends in India for keeping in touch and sharing lot of good stories; espe-
cially Yogesh, Swapnil, Rekha, Nikhil, Bhushan, Vijay and Sandeep.
Finally, my deep and sincere gratitude to my family for their continuous and unparalleled
love, help and support. I am grateful to my wife Utkarsha for moral and emotional support
in the final phase of PhD journey. I am forever indebted to my parents for giving me the
opportunities and experiences that have made me who I am. They selflessly encouraged me
to explore new directions in life and seek my own destiny. This journey would not have been
possible if not for them, and I dedicate this milestone to them.
About the author
Sushil Shamrao Shirsath was born on 1st August 1984 in Jamner, Maharashtra, India. He
completed his Bachelor degree in Chemical Engineering from University of Pune, India in
2006 with first class honours. In 2006, he joined Post Graduate Diploma in "Industrial wa-
ter and waste water treatment" from Vishwakarma Institute of Technology, Pune, India and
honoured with first class. In 2007, he started his Master degree at Institute of Chemical
Technology in Mumbai, India under the supervision of Prof. A.V. Patwardhan. He received
his Master degree in Chemical Engineering with a thesis entitled "Modeling of membrane
reactor" in 2009 with first class honours.
After finishing his Master degree, he joined HyCa Technologies, Mumbai, India as a Senior
Engineer. After a work experience of about 10 months he left the company in August 2010
to pursue his career as a PhD.
In November 2010, he started his PhD research project "Granular Flows Down Inclined
Rotating Chutes: Experimental and Simulation Studies" under the guidance of Prof. Hans
Kuipers, Prof. Herman Clercx and Dr. Johan Padding in the Multiphase Reactors group at
Eindhoven University of Technology at Eindhoven, The Netherlands. The research project
was funded by STW. The results of this research are presented in this dissertation.
167

Granular Flows Down Inclined Rotating Chutes PDF

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Granular Flows Down Inclined Rotating Chutes PDF

Încărcat de

Drepturi de autor:

Formate disponibile

Granular flows down inclined rotating chutes :

experimental and simulation studies

Document status and date:

Please check the document version of this publication:

Take down policy

Download date: 07. Apr. 2019

ter verkrijging van de graad van doctor aan de

Sushil Shamrao Shirsath

geboren te Jamner, Maharashtra, India

voorzitter: prof.dr.ir. J.C. Schouten

The research reported in this thesis was ﬁnancially funded by STW.

Publisher: Gildeprint, Enschede, The Netherlands.

iment zowel de oppervlaktesnelheid en de bedhoogte verkregen kunnen worden, waarmee

2 Experimental study of granular ﬂow down a rotating chute 9

3 Numerical investigation of granular ﬂow down a rotating chute 25

3.2 Simulation model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

4 Cross-validation of 3D-PTV for granular ﬂows down rotating chutes 55

Calculation of surface velocity in simulations . . . . . . . . . . . . . . . . . 69

5 Effect of wall roughness on granular ﬂows down rotating semi-cylindrical chutes 81

6 Simulation study of density segregation down rotating rectangular chutes 111

7 General conclusions and recommendations 133

A Appendix: Single particle experiments 139

About the author 167

1.1 General introduction

1.2 Granular material

1.3 Blast furnace

1.4 Charging mechanism in the blast furnace

1.5 Objective of the thesis

Iron ore Coke

1.6 Experimental approach

1.7 Discrete Element Method simulations

1.8 Outline of the thesis

In detail, chapter 2 presents an experimental study of monodisperse ﬂows through inclined

2.2 Experimental setup and techniques

Dynamic weighing scale

Ultrasonic height sensor

Particle image velocimetry

Table 2.1: Experimental settings and materials for experiments.

Rossby number Froude number

Experimental settings and materials

= 0 rpm = 4 rpm = 8 rpm = 16 rpm

2.3 Results and discussions

Mass ﬂow rate

Particle bed height

0 rpm 4 rpm 8 rpm 16 rpm

0.02 0.02 0.02 0.02

0.01 0.01 0.01 0.01

0.02 0.02 0.02 0.02

0.01 0.01 0.01 0.01

Chute length (m)

Surface particle velocity

Streamwise surface velocity along the length of the chute

= 0 rpm = 4 rpm = 8 rpm = 16 rpm

φ = 30 deg φ = 40 deg φ = 50 deg

Ω = 0 rpm single particle

2.5 2.5 2.5

1.5 1.5 1.5

0.5 0.5 0.5

Chute length (m)

Ω = 0 rpm Ω = 4 rpm Ω = 8 rpm Ω = 16 rpm

Chute width (m)

Streamwise surface velocity along the width of the chute

Spanwise surface velocity along the width of the chute