Gravity Separation of Silica Sands for Value Addition

This article was downloaded by: [Enstinet], [Suzan Ibrahim]On: 29 September 2013, At: 00:32Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Particulate Science and Technology: An InternationalJournalPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/upst20

Gravity Separation of Silica Sands for Value AdditionSuzan S. Ibrahim a , Ali Q. Selim b & Ayman A. Hagrass ca Central Metallurgical R&D Institute , Helwan , Cairo , Egyptb Geology Dept., Faculty of Science , Beni Suef Univ. , Egyptc El-Tebbin Institute for the Metallurgical Studies , Helwan , EgyptAccepted author version posted online: 06 May 2013.

To cite this article: Particulate Science and Technology (2013): Gravity Separation of Silica Sands for Value Addition,Particulate Science and Technology: An International Journal, DOI: 10.1080/02726351.2013.800930

To link to this article: http://dx.doi.org/10.1080/02726351.2013.800930

Disclaimer: This is a version of an unedited manuscript that has been accepted for publication. As a serviceto authors and researchers we are providing this version of the accepted manuscript (AM). Copyediting,typesetting, and review of the resulting proof will be undertaken on this manuscript before final publication ofthe Version of Record (VoR). During production and pre-press, errors may be discovered which could affect thecontent, and all legal disclaimers that apply to the journal relate to this version also.

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 1

Gravity Separation of Silica Sands for Value Addition

Suzan S. Ibrahim1,, Ali Q. Selim2, Ayman A. Hagrass3

1Central Metallurgical R&D Institute, Helwan, Cairo, Egypt 2Geology Dept., Faculty of

Science, Beni Suef Univ., Egypt 3El-Tebbin Institute for the Metallurgical Studies, Helwan, Egypt

Address correspondence to Suzan S. Ibrahim, Central Metallurgical R&D Institute, P.O.

Box 87, Helwan, Cairo, Egypt

Abstract

A representative white sand sample was investigated for glass industry. Complete

characterization of the sample was conducted. Chemical analysis of the sample showed

that iron and alumina oxides reached 0.046% and 0.044%, respectively. Dry sieving was

carried out to reject +0.6 mm and -0.10 mm fractions from the sample. The classified -0.6

+ 0.1 mm product was directed to attrition scrubbing. The effect of pulp density, attrition

impeller speed, attrition time and mode was studied. The attrition sand product was

further subjected to gravity separation using "Wilfley" shaking table. Different working

conditions of table separation i.e. sand feeding rate, stroke length, deck inclination, and

dressing water flow rate were optimized. Results showed that the classified -0.6+0.106

mm sand product contained 0.039% Fe2O3 and 0.041% Al2O3 matched the specifications

for the 4th quality sand for sheet and plate glass industry. However, iron and alumina

oxides contents were further decreased to 0.025% and 0.0164% after the attrition process.

The attrition product accepted as 2nd quality for flint containers and table ware

applications. The final sand product after shaking table contained 0.0180% Fe2O3 and

0.090% Al2O3, was applicable for the 1st quality for optical applications.

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 2

KEYWORDS: silica sand, glass sand, attrition scrubbing, shaking table

1. INTRODUCTION

The term "silica sand" is used for sedimentary white sand ores which have silica as their

main constituent and SiO2 content exceeds 99%, Ariffin (2004). Meanwhile the term

“glass sand” is used when silica sand satisfy certain grain size distribution and certain

iron oxide content in silica sands. The particular application of glass has been classified

based on the quality of silica sand as outlined in the British standard specifications,

Dolley (2006) and the American Standard Specification, Dumont (2003). The

classifications are based on the chemical compositions in terms of the purity of silica (%)

and other elemental constituents. In glass industry, the majority of commercial glass can

be divided into four areas - advanced and specialty glass, fiber glass, container glass and

flat glass, Mustafa et al. (2011), Edler and Domenico (2000), Mills (1983), Mclaws

(1971).

Impurities in glass sands are usually present as free and coated iron oxides, clay, titania

and refractory minerals. The iron is being the most detrimental impurity. It can be

reduced by a number of physical, physico-chemical or chemical techniques. The most

appropriate method depends on the mineralogical forms and distribution of iron in the

ore. While much of the liberated impurities can be reduced or removed by physical

operations such as size separation (screening), gravity separation, magnetic separation

etc., sometimes, physico-chemical (flotation) or even chemical methods (leaching etc.)

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 3

are to be adopted for effective removal of iron which may be in intimate association with

the mineral quite often superficially, Harben (1999), Ay and Erica (2000), Al-Maghrabi

(2004), Sundararajan et al (2009).

The aim of this study is to try to adopt simple, cost effective as well as environment-

friendly processes and operations for value addition of silica sand sample for different

glass applications.

2. EXPERIMENTAL

2.1. Materials& Characterization Equipments

Representative technological white sand sample from Abu Zeneima locality was kindly

supplied by Sinai Manganese Co. The sand sample was thoroughly directed to mixing

and splitting preparation. Wet/dry classification by sieving was applied using “Fritsch,

Analysette”, Germany and “Retsch”, F. Kurt Retsch, GmbH& Co. kG, Haan, Germany,

respectively. Complete chemical analysis of the original sample was carried out using

“Panalytical-Axios” XRF unit while routine analysis for iron and alumina oxides

determination were followed up using “Perkin-Elmer Analyst 200” atomic absorption.

2.2. Procedures

The original sand sample was directed to dry classification by sieving to reject +0.60 mm

and – 0.10 mm fractions. The classified sand was evaluated and the rejected fractions

were petrography examined. The classified sample -0.60 + 0.106 mm was subjected to

attrition scrubbing operation using a Perspex container of a laboratory “Denver “flotation

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 4

cell, Figure 1 [A] and a highly efficient assembled impeller system, Figure 1 [B] were

used in the attrition process. Optimization of the attrition process was applied by studying

the attrition pulp solid %, attrition tank impeller speed, and attrition time and mode. The

scrubbed sand was deslimed and washed with water very well at 106 microns sieve

before any new attrition process. The attrition sand was evaluated and the rejected

fraction attrition was petrography examined. The optimized attrition sand product was

further directed to gravity separation using "Wilfley" shaking table. The table surface

dimensions were 3000 mm length x 875mm width with a nominal deck area 4.0 m2.

Throughout the separation experiments, the sand samples were kept in suspension by

manual stirring. The working parameters affect the separation efficiency like stroke

length, deck angle, wash water flow rate, and feeding rate were optimized. The position

of splitter plates was noticed thoroughly through all separation tests to control the

separated sand products in terms of quantity and quality. Different rejected impurities

after each beneficiation technique were petrography investigated using polarizing

microscope.

3. RESULTS& DISCUSSIONS

3.1. Characterization Of The Original Sand Sample

Table 1 illustrated the chemical analysis of the original white sand sample. From Table 1,

it was shown that the sample contained 99.44% SiO2, 0.046% Fe2O3, and 0.044% Al2O3.

According to chemical specifications, the present original sample was categorized as high

grade silica sand, Mclaws (1971). Dry screen analysis of the sample was illustrated in

Figure 2. High iron oxide content in the -0.106 mm fraction was noticed, Table 2. Sieve

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 5

classification of the sample showed remarkable improvement in the sample grade after

the rejection of +0.6 mm and -0.106 mm fractions. Iron and alumina oxides reached in

the classified sample 0.039% and 0.041% from 0.046% Fe2O3 and 0.044 Al2O3,

respectively. Figure 3 showed the petrography pictures of rejected minerals after

screening. The +0.6 mm coarse silica particulates contaminated with magnetite inclusions

was shown in Figure 3 [A]. Fine magnetite, well developed zircon crystals, epidote,

muscovite, tourmaline, and clinozoisite minerals were shown in Figure 3 [B], [C], [D],

and [E], respectively. Dry classified sand product matched the criteria for the 4th quality

for sheet and plate glass application, Dolley (2006), Dumont (2003).

3.2. Attrition Scrubbing Of The Classified Sample

3.2.1. Effect Of Pulp Density

Figure 4 illustrated the effect of solid % of the attrition pulp on the iron oxide content of

the sand product. It was found that at 65-70% solid, satisfied particle-to-particles

scrubbing took place. Figure 5 showed the aggressive turbulence motion took place

inside the attrition tank. At these conditions, the pulp viscosity of the slurry was just

proper enough to allow the slurry to move freely in the attrition tank. It was noticed that

at diluted attrition pulp, i.e. lower than 65 %, there was sufficient water in the slurry to

allow the particles to stay relatively apart and prevent the necessary particle-to-particle

contact required to scrub the ferruginous coatings from the sand grains surfaces. In case

of thick pulps, i.e. above 75% solid, the slurry became too viscous to allow the required

particle-particle friction, while the attrition tank impellers showed difficulty to move the

slurry sufficiently. At these attrition conditions, the removal efficiency of iron and

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 6

aluminum oxides showed remarkable deterioration, Figure 4. It is worthy to mention that

in order to process the high percent solids slurry on a consistent basis, the power

transmission system employed in the attrition system is extremely critical. The best

possible solution is a gear driven attrition machine. Although v-belt systems tend to work

fine when the attrition units are new, as they age and are not properly maintained,

slippage occurs and the impeller will not rotate with the resistance caused by the high

percent solids. At this point, operators tend to add additional water to lower the percent

solids and the scrubbing becomes ineffective, Wills and Napier-Munn (2006).

3.2.2. Effect Of Attrition Time& Mode

Figure 6 illustrated the effect of attrition time on the efficiency of the attrition scrubbing

of the classified sand. It was shown that gradual improvement in sand grade with

increasing attrition time, Figure 6. Results showed that about 28% and 46% of the total

iron and aluminum oxides contents were removed after attrition for 30 min, respectively.

By increasing the attrition time to 60 min, no remarkable improvement was noticed,

Figure 6.

It was noticed that gradual decrease in iron and alumina oxides contents to 295 and 220

ppm in case of one attrition step for 30 min, 281 and 179 ppm in case of two attrition

steps (each step 15 min), and to 254 ppm and 164 ppm in case of three attrition steps

(each step 10 min) for Fe2O3 and Al2O3 contents, respectively. This behavior remarked

that the removing efficiency of these harmful oxides increased from 26.6% and 46% after

one attrition step for 30 min, to 30% and 55.9%, when the sample was attrition in two

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 7

steps and increased to 37.6% and 59.8% when the attrition was conduct as three separate

steps for Fe2O3 and Al2O3, respectively. These results were recorded, providing that

proper desliming on 0.1 mm screen followed by thorough water washing of the attrition

product must carried out prior the next new attrition step.

Generally, it could be concluded that two to three shorter scrubbing times steps in

between is more effective than one long scrubbing time. The number of attrition

scrubbing cells in the circuit was important. Although one cell may have sufficient

volume to meet the retention time needed, one cell will result in significant short-

circuiting of the feed material. That was, if the average particle retention time is 5

minutes, for example; there may be particles that will only have retention times of 2

minutes and others at 8-10 minutes. Accordingly, it is important that at least two attrition

scrubber cells have to be used in the attrition circuit to provide the necessary retention

time. Three or four cells were preferred whenever possible, Wills and Napier-Munn

(2006).

3.2.3. Effect Of Attrition Impeller Speed

Figure 7 illustrated the effect of machine impeller speed on the attrition process. It was

shown that gradual improvement in sand grade with increasing impeller speed, Figure 7.

By decreasing the impeller speed to less than 2400 rpm during the attrition process, a

great loss in the attrition efficiency occurred. By increasing the impeller speed from 2400

rpm to 2700 rpm, no remarkable improvement in attrition efficiency was noticed, Figure

7.

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 8

Petrography pictures of the rejected fractions after the multi-stage attrition process at

optimum conditions were illustrated in Figure 8. It was clear that magnetite inclusions

were dominant either on sand surface or on other minerals surfaces like biotite,

muscovite, staurolite, and ziosite, Figure 8.

Figure 9 illustrated a comparative block diagram with respect to Fe2O3 and Al2O3

contents in original, classified, and attrition sand products. The chemical specifications of

the attrition sand producr matched the 2nd quality glass sand for flint containers and table

ware.

3.3. Shaking Table Separation Of Attrition Sands

In shaking table separation, the feeding ore was subjected to two forces, that due to the

table motion and that at right angles to it due to the flowing film of water. The net effect

was that the particles move diagonally across the deck from the feed end. Since the effect

of the flowing film depended on the size and density of the particles, they fanned out on

the table, the smaller, denser particles were tiding highest towards the concentrate launder

at the far end, while the larger lighter particles were washed into the tailings launder,

which was run along the length of the table, Burt and Mills (1984).

3.3.1. Effect Of Feeding Rate

It was remarked that at feeding rate up 0.75 kg/min (45 kg/hr), sand product of minimum

iron oxide reached 180 ppm from a feeding sample contained 254 ppm was obtained. At

feeding rates above 0.75 kg/min, it seemed that the space between the table cleats became

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 9

clogged by sand and further dilation of the bed was hindered. This allowed the escape of

entered fine heavy minerals with the cross water stream. Although true flowing film

concentration required a single layer of feed, in practice a multilayered feed was

introduced on to the table, enabling much larger tonnages to be dealt with. Important to

clarify that vertical stratification due to shaking action took place behind the riffles,

which generally was run parallel with the long axis of the table and were tapered from a

maximum height on the feed side, till they died out near the opposite side, where part of

which was left smooth. In the protected pockets behind the riffles the particles stratified

so that the finest and heaviest particles were at the bottom and the coarsest and lightest

particles were at the top. Layers of particles were moved across the riffles by the

crowding action of new feed and by the flowing film of wash water. Due to the taper of

the riffles, progressively finer sized and higher density particles were continuously being

brought into contact with the flowing film of water that topped the riffles. Final

concentration took place at the un-riffled area at the end of the deck, where the layer of

material was at this stage usually only one or two particles deep, Wills and Napier-Munn

(2006).

3.3.2. Effect Of Motor Speed And Stroke Length

Figure 10 illustrated the effect of stroke length of the table motion on the separation

efficiency of the attrition sands. Results showed that at stroke length 2.5cm (25mm) and

motor speed 300 rpm, the maximum separation efficiency of iron oxides was recorded.

The iron oxide content of the product assayed 181 ppm from feeding sand contained 254

ppm Fe2O3 with removal % reached 28.40, Figure 10. By the increase in motion speed,

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 10

the deck surface went forward until it was jerked to a halt before being sharply reversed,

allowing the particles to slide forward during most of the backward stroke due to their

built-up momentum. Generally, low speed and long stroke shaking were being suitable

for coarse feeds and the reverse for fine feeds, Burt and Mills (1984).

3.3.3. Effect Of Deck Inclination

Figure 11 illustrated the effect of deck inclination or tilt slope on the separation process.

On a horizontal deck (90°), there is no motion of particles. As the deck was tilted,

particles begin to move until all of them were moving. Results that were shown in Figure

11 indicated that at 7.5o-8° (from the horizontal) a clean sand product was obtained

having 181 ppm Fe2O3 with removal % reached 28.7. By increasing the deck tilt more

than 7.5o, notable decrease in sand quality was decreased from 181 ppm to 205 ppm,

Figure 11. This could be explained when the deck slope increased, the transporting power

of particles increased and narrowed the bands of the products at the concentrate end and

made accurate splitting difficult.

The shaking table slopes in two directions, across the riffles from the feed to the tailings

discharge end and along the line of motion parallel to the riffles from the feed end to the

concentrate end. The latter was greatly improved separation due to the ability of heavy

particles to “climb” a moderate slope in response to the shaking motion of the deck. The

elevation difference parallel to the riffles should never be less than the taper of the riffles;

otherwise wash water tended to flow along the riffles rather than across them. It was

stated that the effect of deck inclination or deck tilt has only limited effects on the

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 11

separation mechanism, especially within the riffles. This parameter essentially affects the

transport of the particles across the deck. Ideally, the side tilt was set at the minimum at

which a good distribution of material can be obtained on the deck. High tilt caused more

material to discharge along the lights discharging end, Wills and Mills (1984).

3.3.4. Effect Of Water Flow Rate

It is stated that the water flow and transverse slope of the deck are interdependent and

both are dependent on the size of feed, Burt and Mills (1984). The requirements for good

separation are that, solids should settle in the riffles, the pulp should be sufficiently fluid

to allow stratification and that there should be sufficient velocity of cross water flow to

carry off the upper strata as the riffle support is withdrawn. In this study, the quantity of

water flowed over the table was varied from a few liters to almost 40 liter per min. It was

found that the most suitable wash water flow rate was 15 L/min. At this condition, the

iron content of the separated clean sand assayed 181 ppm from 254 ppm Fe2O3 in the

feeding sample. By increasing the rate of water above 15 l/min, some of the heavy

mineral particles washed down with the table lights and lowered the sand product quality.

On the other hand, at water flow rate below 20 l/min, the flowing water was not sufficient

to form a freely moving film on the deck deep enough to cover the relatively large

particles.

Figure 12 showed the petrography picture of rejected minerals after shaking table

operation. Magnetite, rutile, and zircon were the major harmful minerals removed after

tabling separation, Figure 12. Their difference in size and shape with that of silica sand,

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 12

as well as a concentration criterion of 2 or more showed reasonable separation via

shaking table. The sand product after shaking table matched the chemical criteria for the

1st quality glass sand for optical application. Figure 13 showed the effect of applied

upgrading techniques on removing iron oxide from the sand sample under investigation.

4. CONCLUSION

The present study was carried out to evaluate its beneficability for value addition.

Petrography characterization showed that the main impurities in the sample are iron

bearing minerals (like magnetite, biotite, oivine, and staurolite), alumina bearing minerals

like tourmaline, clinozoisite, muscovite, zoisite, and epidote, and rutile mineral as the

main titanium bearing mineral. Zircon mineral also was found as the main refractory

mineral. Results revealed that dry screening to remove +0.60 and -0.10 mm fractions

upgraded the sand sample to give 4th quality grade suitable for sheet and plate glass.

Attrition scrubbing of the classified sand was applied at the optimum to give a product

matched the chemical specifications of the 2nd quality glass sand for flint containers and

table ware. When the attrition sand product was further subjected to shaking table at the

optimum separating conditions a sand product matched the 1st quality glass sand grade

for optical applications was produced. The present study succeeded to adopt simple, cost

effective as well as environment-friendly processes and operations for value addition of

silica sand from Abu Zeneima locality, south west of Sinai Peninsula, Egypt.

REFERENCES

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 13

Al-Maghrabi, M. 2004. Improvement of low-grade silica sand deposits in Jeddah area.

JKAU, Eng. Sci. 15(2): 113-128.

Ariffin, K. 2004. What is Silica?. EBS 425 – Mineral Perindustrian.

Ay, N. and E. Arica. 2000. Refining Istanbul’s silica sand. American Ceramic Society

Bulletin. 79(8).

Burt, B., and C. Mills. 1984. Gravity concentration technology. New York: Elsevier.

Dolley, T. 2006. Mineral commodity summaries. In U.S. geological survey. pp 144-145.

Dumont, M. 2003. Silica / quartz. In Canadian Minerals Year book.

Elder, J. and J. Domenico. 2000. The role of physical processing enhancing the quality of

industrial minerals. In 14th Industrial Minerals International Congress. Denver,

Colorado, USA.

Harben, W. 1999. Silica and quartz. In The nIdustrial Minerals HandyBook, 3rd Edition,

Industrial Minerals Information Ltd. Surrey, UK. pp. 183-190.

Mclaws, I. 1971. Uses and specifications of silica sands. In Research Council of Alberta,

Edmonton, Alberta 71(4).

Mills, N. 1983. Glass Raw Materials, Industrial Minerals and Rocks, 5th ed. The

American Institute of Mining, Metallurgical, and Petroleum Engineers. Inc., NY. pp.

763-772.

Mustafa, A., N. Bader, T. Khachiek, I. Fleah, I. Issa. 2011. Bio-beneficiation of silica

sands for crystal glass industry from Ardhuma location, Iraqi western desert. Iraqi

Bulletin of Geology and Mining 7 (1): 77-86.

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 14

Sundararajan, M., S. Ramaswamy, P. Raghavan. 2009. Evaluation for the Beneficiability

of White Silica Sands from the Overburden of Lignite Mine situated in Rajpardi district

of Gujarat, India. Minerals & Materials Characterization & Engineering 8(9): 701-713.

Wills, B. and T. Napier-Munn. 2006. Mineral processing technology, an introduction to

the practical aspects of ore treatment and mineral recovery, Elsevier Science &

Technology.

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 15

Table 1. Complete Chemical Analysis of Original Sand Sample

Constituent Wt.% Constituent Wt.%

SiO2 99.439 Na2O 0.010

Al2O3 0.044 K2O 0.010

Fe2O3 0.046 P2O5 0.010

TiO2 0.030 Cl 0.010

MnO 0.011 SO3 0.010

CaO 0.060 L.O.I 0.300

MgO 0.020

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 16

Table 2. Dry Size/Chemical Analysis of Original Sand Sample

Size Fraction, mm Cum. Wt% Retained Fe2O3% Al2O3%

+0.60 3.18 0.049 0.014

-0.60+0.42 39.13 0.037 0.038

-0.42+0. 20 92.33 0.026 0.042

-0.20+0.106 99.00 0.165 0.045

-0.106 100.00 0.590 0.380

Total 0.046 0.049

Original sample 0.046 0.044

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 17

Figure. 1. [A] Denver flotation cell and [B] assembled impeller system

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 18

Figure. 2. Dry particle size distribution of original sand sample

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 19

Figure. 3. Petrography pictures show [A] aggregate of opaque grains showed some

degrees of replacement by light minerals and silica grains, C.N. [B] aggregate of opaque

grains and some of them showed different degrees of replacement by light minerals, C.N.

[C] well developed zircon crystals, epidote, and muscovite, C.N., [D] biotite and

tourmaline minerals, PPL,. [E] clinozoisite mineral, CN.

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 20

Figure. 4. Effect of pulp solid % on the attrition process

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 21

Figure. 5. Strong turbulence inside the attrition tank

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 22

Figure. 6. Effect of attrition time on the attrition efficiency

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 23

Figure. 7. Effect of Attrition impeller speed on the attrition efficiency

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 24

Figure. 8. Petrography pictures show [A] aggregate of biotite, quartz and magnetite PPL

[B] Replacement of magnetite by muscovite, PPL [C] relicts of magnetite in muscovite

mineral, CN. [D] staurolite, zircon, muscovite and magnetite,CN. [E] magnetite

inclusions within biotite grain, PPL, [F] zoisite mineral grain, CN.

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 25

Figure. 9. Fe2O3 and Al2O3 contents in the original, classified, and attrition products

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 26

Figure. 10. Effect of stroke length on Fe2O3 removal

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 27

Figure. 11. Effect of table deck inclination on the product Fe2O3 content

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 28

Figure. 12. [A] Equant magnetite particle, PPL, 10x., [B] equant rutile grain, CN., 30x,

[C] rutile grain with typical red color, PPL, 25 x, [D] well-developed rutile needle, PPL ,

25x, [E] biotite and rutile grains, PPL, 20x., [F] magnetite inclusion within the zircon

crystal, CN, 100x.

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 29

Figure. 13. Iron oxide contents in sand products after different beneficiation techniques

Dow

nloa

ded

by [

Ens

tinet

], [

Suza

n Ib

rahi

m]

at 0

0:32

29

Sept

embe

r 20

13