Investigations on the influence of dextran during beet ... · 4.6 Elucidation of crystallization...

$: Investigations on the influence of dextran during beet ... · 4.6 Elucidation of crystallization kinetics in presence of dextran molecules.. 72 4.7 Influence of dextran molecule fractions$
Investigations on the influence of dextran during beet sugar

production with special focus on crystal growth and morphology

Vorgelegt von

M.Sc.

El-Sayed Ali Abdel-Rahman Soliman

aus Sohag (Ägypten)

Von der Fakultät III - Prozesswissenschaften

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktor der Ingenieurwissenschaften

-Dr.-Ing.-

genehmigte Dissertation

Promotionsausschuss:

Vorsitzender: Prof. Dr. rer. nat. F. Behrendt

Berichter: Prof. Dr. -Ing. T. Kurz

Berichter: PD Dr.-Ing. R. Schick

Berichter: Assoc. Prof. Dr. S. Ahmed

Tag der wissenschaftlichen Aussprache: 13.08.2007

Berlin 2007

D 83

This work is dedicated to:

The memory of my parents

My wife Asmaa

Contents

I

ACKNOWLEDGMENT

“My great thanks to ALLAH for all gifts he gave me’’

In the first place, I would like to thank in the first place my supervisor Prof. Dr. T.

Kurz, for many decisive scientific contributions, continued support, his

recommendations, motivations, countless discussions and for his helpful guidance

and valuable advices. He gave me the opportunity to extend my scientific education

at numerous of international conferences. Special thanks also go to Dr .R. Schick, for

his valuable advice, encouragement and scientific contributions. I would like to thank

Prof. Dr. Fleischer, for supporting this work. I can not forget to extend my thanks to

Prof. Dr. Ezzat Abdalla, president of Assiut University. I gratefully acknowledge

Prof. Dr. S. El-Syiad for his advice and encouragement. I could never have

embarked and started all of this without his prior teaching in sugar technology. Thus

he opened up an unknown area to me. I am gratefully thanking Prof. Dr. I. Makkawy

for his help in statistical analyses of this work. I thank Dr. S. A. Ahmed for his helps

in this work. My deep thanks go to Dr. Q. Smejkal, for his scientific contributions in

the progress of the work. Also, thanks are given to all staff members of the

Department of Food Process Engineering, Technische Universität Berlin, who gave

me hands during the progress of the work, in particular, Dr. B. Seidel, Dr. K.

Ruprecht, S. Gramdorf, S. Jährig, S. Brodkorb, A. Bagherzadeh, H. Rudolph, A.

Hentschel, Y. Pierre, Dr. F. Idler, P. Seifert, R. Krämer, A. Gebauer, M. Fischer

and R. Groß. I gratefully acknowledge Jülich sugar factory, especially, Dr. M.

Lorenz, Aarberg sugar factory, especially, T. Frankenfeld and Delta sugar factories

especially H. Gad. My special thanks go to my wife Asmaa for her support and

comfortable atmosphere during this work. I am extremely grateful to the Egyptian

Government and Missions Office for the financial support during my studies in

Germany, especially, Prof. Dr. Galal Elgemeie.

Finally I am deeply thank all staff members, my colleagues and workers of the Food

Science and Technology Department, Faculty of Agriculture, Assiut University,

Egypt for their continuous encouragements.

Contents

II

CONTENTS

ACKNOWLEDGMENT ............................................................................................I

CONTENTS .............................................................................................................. II

LIST OF FIGURES...................................................................................................V

LIST OF TABLES................................................................................................... IX

1 Introduction and outline ................................................................................... 1

2 Review of literature ........................................................................................... 5

2.1 Structure of dextran ..................................................................................... 5

2.2 Microbial loading in sugar factories ............................................................ 7

2.3 The common methods of dextran fractions determination ........................ 11

2.4 Dextran content during the process of sugar production ........................... 13

2.5 Dextrans associated with processing problems ......................................... 16

2.6 Crystallization process............................................................................... 19

2.6.1 Growth rate of sucrose crystals.......................................................... 19

2.6.2 Crystallization kinetics ...................................................................... 22

2.6.3 Parameters influencing crystallization kinetics ................................. 27

2.6.4 Crystal morphology ........................................................................... 29

2.7 The Economic gain .................................................................................... 32

3 Material and methods...................................................................................... 36

3.1 Material ...................................................................................................... 36

3.2 Analytical methods .................................................................................... 36

3.2.1 Determination of dextran ................................................................... 36

3.2.1.1 Robert method................................................................................ 36

3.2.1.2 Haze method .................................................................................. 37

3.2.2 Microbiological experiments ............................................................. 37

3.2.2.1 Isolation ......................................................................................... 37

3.2.2.2 Identification.................................................................................. 37

3.2.2.2.1 Gas and acid formation ............................................................ 37

3.2.2.2.2 Catalase test ............................................................................. 38

3.2.2.2.3 Gram characteristics (KOH-Test) ............................................ 38

Contents

III

3.2.2.2.4 Identification by API 50 CHL test ........................................... 38

3.2.2.2.5 L/D-Lactic acid test ................................................................. 39

3.2.3 Crystallization experiments ............................................................... 39

3.2.3.1 Measurement of growth rate of sucrose crystals ........................... 39

3.2.3.1.1 Required amount of dextran and seed...................................... 40

3.2.3.1.2 Calculation of the growth rate of sucrose crystals:.................. 41

3.2.3.2 Dynamic viscosity.......................................................................... 42

3.2.3.3 Crystal morphology and surface topography................................. 42

3.2.3.4 Image analysis................................................................................ 42

3.2.4 Statistical analysis.............................................................................. 43

4 Results and discussion ..................................................................................... 44

4.1 Sensitivity and accuracy of different methods for the determination of

dextrans of varying molecular mass ...................................................................... 44

4.1.1 Robert’s Copper method sensitivity .............................................. 44

4.1.2 Haze method sensitivity................................................................. 48

4.2 Microbial sources of dextran an identification of relevant microorganisms

in sugar factories.................................................................................................... 51

4.3 Levels of dextran contents in different sugar beet factories ...................... 55

4.4 Quality of factory final products and their relationship to the levels of

dextran during different industrial periods ............................................................ 58

4.5 Influence of dextran concentrations and molecular fractions on the rate of

sucrose crystallization in pure sucrose solutions ................................................... 63

4.5.1 Influence of different temperatures on growth rate of sucrose crystals

in the presence of dextran .................................................................................. 69

4.6 Elucidation of crystallization kinetics in presence of dextran molecules.. 72

4.7 Influence of dextran molecule fractions on sucrose solution viscosity ..... 74

4.8 Influence of dextran on the morphology and surface topography of sucrose

crystals in presence of dextran............................................................................... 76

4.8.1 Crystal morphology ........................................................................... 76

4.8.2 Surface topography ............................................................................ 77

4.9 Technical and technological consequences and future perspectives ......... 86

Contents

IV

5 Summary........................................................................................................... 90

6 References......................................................................................................... 94

7 Appendix......................................................................................................... 102

8 C. V. and List of Publications ....................................................................... 106

List of figures

V

LIST OF FIGURES

Figure 1: Dextran formed from sucrose by Leuconostoc m. ........................................ 5

Figure 2: Chemical structure of dextrans ..................................................................... 6

Figure 3: Micrographs of four strains of Leuconostoc spp by scanning electron

microscopy (SCIMAT, 2006) ............................................................................... 9

Figure 4: Growth curves of Leuconostoc mesenteroides subsp. mesenteroides MCRI

1 in MRS broth at 30-40°C (Hamasaki et al., 2003) (30°C (•), 35°C ( ), 37°C

( ), and 40°C (X). ABS, absorbance)................................................................. 10

Figure 5: The major steps for the manufacture of sugar from sugar beets (Südzucker

AG website, Germany) ...................................................................................... 15

Figure 6: Growth rate at 60 °C as a function of supersaturation at different purities,

(Schliephake and Ekelhof, 1983)........................................................................ 21

Figure 7: Crystal growth rate as a function of temperature at different purities and

constant supersaturation (Schliephake and Ekelhof, 1983)................................ 21

Figure 8: Surface reaction coefficient kR dependent upon the reciprocal value of the

absolute temperature T given for different purities (Ekelhof, 1997).................. 25

Figure 9: Hypothetical crystal (a) with schematic drawing of the three kinds of faces

PBCs (b) ............................................................................................................. 26

Figure 10: Diagrammatic representation of the processes involved in crystal growth

from solution, according to Elwell and Scheel ,(1975) as modified by Kruse and

Ulrich, (1993)..................................................................................................... 27

Figure 11: The viscosity of the pure and impure sucrose solution according to

Ekelhof, (1997)................................................................................................... 29

Figure 12: Schematic representation of a sucrose crystal with the surface designation

according to Vavrinecz ( 1965) .......................................................................... 30

Figure 13: (A) Sucrose crystal growth in the presence of dextran (40g/100g water),

(B) One of the crystals shown in (A) and (C) Sucrose crystal growth in the

presence of glucose, fructose and dextran (50g/100g water each) (According to

Bubnik et al., (1992)........................................................................................... 31

Figure 14: Partitions of sugar loss during different process stages of sugar production

(Ekelhof, 1997)................................................................................................... 33

Figure 15: Colony of bacteria mixed in a 5% solution of H2O2 demonstrating a

positive and negative test for producing of catalase. ......................................... 38

List of figures

VI

Figure 16: Pilot scheme of laboratory crystallization device..................................... 40

Figure 17: Standard Curves of dextran T40, T500 and T2000 by copper complex

method................................................................................................................ 45

Figure 18: Sensitivity of copper complex method to determine dextran T40 in

sucrose solutions ................................................................................................ 47





Figure 21: Sensitivity of Haze method to determine dextran T40 in sucrose solutions

............................................................................................................................ 49

Figure 22: Sensitivity of Haze method to determine dextran T500 in sucrose

solutions ............................................................................................................. 49

Figure 23: Sensitivity of haze method to determine dextran T2000 in sucrose

solutions ............................................................................................................. 50

Figure 24: Deteriorated sugar beet after harvesting................................................... 52

Figure 25: Microorganisms growth in sugar beet raw juice at room temperature ..... 53

Figure 26: Scanning electron micrograph of the isolated Leuconostoc mesenteroides.

............................................................................................................................ 55

Figure 27: Dextran levels at 3 Egyptian factories (F1, F2 and F3) during the

production processes .......................................................................................... 56

Figure 28: Dextran levels in final products (white sugar and molasses) of 3 Egyptian

factories (F1, F2 and F3).................................................................................... 57

Figure 29: The relationship between the three factories based on cluster analysis of

some characters (Single linkage squared Euclidean distance)........................... 58

Figure 30: The relationship between molasses purity and the industrial periods. ..... 59

Figure 31: Influence of the industrial period on the dextran amount during production

processes ............................................................................................................ 60

Figure 32: Dextran levels in sugar and molasses production of different industrial

periods ................................................................................................................ 61

Figure 33: The relationship between the dextran content in syrup and different

produced sugars.................................................................................................. 62

Figure 34: The partitions of dextran during different crystallization stages at different

industrial periods................................................................................................ 62

List of figures

VII

Figure 35: Dry substance content of mother liquor in isothermal crystallization

experiments at 60 °C in pure sucrose solution and after addition of 500, 1500

and 2000 mg/kg DS of dextran T40................................................................... 64

Figure 36: Growth rate of sucrose crystals in isothermal crystallization experiments

at 60 °C in pure sucrose solution and after addition of 500, 1500 and 2000

mg/kg DS of dextran T40................................................................................... 64


experiments at 60 °C in pure sucrose solution and after addition of 500, 1500

and 2000 mg/kg DS of dextran T500................................................................. 65


at 60 °C in pure sucrose solution and after addition of 500, 1500 and 2000

mg/kg DS of dextran T500................................................................................. 66

Figure 39: Dry substance contents of mother liquor in isothermal crystallization

experiments at 60°C in pure sucrose solutions and after addition of 500, 1500

and 5000 mg/kg DS of dextran T2000............................................................... 67

Figure 40: Growth rates of sucrose crystals in isothermal crystallization experiments

at 60°C in pure sucrose solutions and after addition of 500 - 5000 mg/kg DS of

dextran T2000 .................................................................................................... 67


experiments at 60 °C in pure sucrose solution and after addition of 1500 mg/kg

DS of dextran T40, T500 and T2000 ................................................................. 68


at 60 °C in pure sucrose solution and after addition of 1500 mg/kg DS of

dextran T40, T500 and T2000............................................................................ 69

Figure 43: Dry substance content at 70, 65 and 60°C in control and in presence of

dextranT2000 (1500 and 5000ppm) during crystallization process................... 70

Figure 44: Growth rate of sucrose crystals at 60, 65 and 70 °C after addition of 1500

mg/kg DS of dextran T2000............................................................................... 71

Figure 45: Growth rate of sucrose crystals at 60 and 70 °C after addition of 5000

mg/kg DS of dextran T2000............................................................................... 71

Figure 46: The scheme of the two processes of diffusion and surface reaction. 2λ=

gap between crystals, λ = half distance between crystals, δ = thickness of the

boundary layer, kD = diffusion coefficient, kR = surface reaction, NS = non-sugar,

List of figures

VIII

CL, CG, CSat are the sucrose concentrations of liquor, boundary layer and

saturation at the crystal-solution interface. ........................................................ 73

Figure 47: Dextran adsorption on impure particles.................................................... 73

Figure 48: Influence of addition of 1500 mg/kg DS of T40, T500 and T2000 dextran

on the dynamic viscosity of a 60% sucrose solution between 50 and 70 °C ..... 74

Figure 49: Influence of different dextran concentrations (500, 1500, 2500, 5000 ppm)

on the dynamic viscosity at different temperatures (50-70°C) .......................... 75

Figure 50: The crystal surface of a sucrose crystal during the drying process

(according to Bunert und Bruhns 1995)............................................................. 76

Figure 51: Effect of dextran on morphology of sucrose crystals, a) Normal crystal, b)

Crystal grown in presence of dextran (needle crystal) and c) Impure particles on

crystal surface (crystal grown in technical sucrose solution)............................. 77

Figure 52: An example of influences of dextran T40 on sucrose crystal shape during

the crystallization process .................................................................................. 77

Figure 53: Surface topography of a laboratory sucrose crystal grown in absence of

dextran at 60°C................................................................................................... 78

Figure 54: Surface topography of sucrose crystal grown in the presence of 1500 mg

dextran T40 per kg DS at 60 °C......................................................................... 79

Figure 55: Surface topography of a sucrose crystal grown in the presence of 5000 mg

dextran T2000 per kg DS at 60°C...................................................................... 80

Figure 56: Effect of crystallization temperature on the surface topography in the

presence of dextran ............................................................................................ 81

Figure 57: Turbidity phenomena in many European sugar beet factories (a, b) during

the thick juice campaign..................................................................................... 82

Figure 58: The magnification image of Figure 57 ..................................................... 83

Figure 59: After product sugar crystal in one of European sugar beet factory during

the thick juice campaign..................................................................................... 83

Figure 60: 3-product scheme of crystallisation.......................................................... 85

Figure 61: Influence of dextran on many crystallization parameters during

crystallization process. ....................................................................................... 86

Figure 62: The relationship between dextran additions and crystallization time during

the crystallization process .................................................................................. 87

List of Tables

IX

LIST OF TABLES

Table 1: The linkage structures of dextran................................................................... 7

Table 2: Sucrose consumption and dextran production at different time intervals,

Leuconostoc mesenteroides 3A; culture medium 10% sucrose basal medium;

incubation temperature 30 °C (Guglielmone et al., 2000) ................................. 10

Table 3: Dextran in process streams measured at different Louisiana sugar

factories(F1, F2, F3 and F4) (Cuddihy and Day, 1999)..................................... 16

Table 4: Summary of the detrimental effects of dextran in terms of the losses it leads

to according to Singleton et al., (2001). ............................................................. 18

Table 5: Effects of dextran in juice on molasses purity ............................................. 34

Table 6: Dextran levels and sucrose loss in raw sugars (Bose and Singh, 1981) ...... 34

Table 7: Conclusion of evaluation of Roberts and Haze methods to determine

different dextran molecules................................................................................ 51

Introduction and outline

1

1 Introduction and outline

More than 100 countries produce sugar, 74 % of which is made from sugar cane

grown primarily in the tropical and sub-tropical zones of the southern hemisphere,

and the balance from sugar beet which is grown mainly in the temperate zones of the

northern hemisphere (SIL, 2007).

The world sugar production in 2005/06 and 2006/07 was estimated as 152.710 and

160.203 (million tons, raw value), 79 % of which is produced by the world's top ten

sugar producers. On the other hand the world consumption was 149.782 and 153.008

(million tons, raw value) in 2005/06 and 2006/07, respectively (Gudoshnikov, 2007).

In Egypt, total sugar production increased to 1.675 million tons a year in 2006/07

from 1.575 million tons in 2005/06, but consumption soared to over 2.5 million tons

a year (Gudoshnikov, 2007).

Sugar cane is the second most important crop in Egypt. Since the first sugar factory

was established in 1818, there are eight factories for raw and plantation white sugar

production from sugar cane as well as one refinery concentrated in middle and upper

Egypt (Khaleifah, 2001). In addition, there are four sugar beet factories in the north

of Egypt. There is an intention in the policy of the Egyptian government to expand

the cultivation of sugar beet in the new land in the north, consequently, the number

of beet sugar factories is increasing.

Regarding beet sugar production, differences between theoretical and practical

quality during sugar extraction from sugar beet and an increase of molasses purity

have been observed in the Egyptian sugar beet factories and in many another

countries, especially at the end of the industrial season. Climatic conditions and a

long time from harvesting to manufacturing cause a drop of sugar beet quality. Also

freezing and thawing cause considerable changes in the chemical composition and

thus processability of sugar beet (Kenter and Hoffmann, 2006). This is of particular

interest also for the European sugar industry as the European sugar factories intend a

prolongation of the industrial campaign to the freezing period.


2

In this context bacterial polysaccharides have been recognized as a serious problem

in sugar processing for more than 100 years (Scheibler, 1874). Since the early years

of sugar manufacturing, there have been many reports especially on the formation of

dextran, the main portion of bacterial polysaccharides during sugar processing, in

sugar factories and refineries.

Dextran fractions occurring in sugar beet and cane are almost entirely produced by

Leuconostoc mesenteroides, a ubiquitous bacterium especially prevalent in any sugar

field. The exact structure of each type of dextran depends on its specific microbial

strain of origin. During storage, several changes in beet quality may occur which

depend on the storage conditions such as duration and temperature. The presence of

dextran in the sugar factories leads to a falsely high polarization, increased viscosity,

lower evaporation rates, elongated crystals (needle grain) and increase of sugar loss

to molasses. However, the most damaging effects of high dextran concentrations in a

technical sucrose solution are foreseen in the crystallization. Dextran slows down the

crystallization rate or even inhibits crystallization (i.e., they have a high melassigenic

effect) (Clarke et al., 1997). Dextrans occurring during sugar production mainly

affect the growth rate and the shape of sucrose crystals during the crystallization

process. Decreasing growth rates and needle-like crystal growth cause relevant

economic and qualitative losses.

Many investigations suggested a minimization of dextran levels in the sugar factory

by controlling of microorganisms from beet harvesting to juice extraction. More

recent approaches propose the application of dextranase enzyme for the removal of

preformed dextran. However, these approaches were not practicable or uneconomical

(Day, 1984; Eggleston and Monge, 2005; Gupta and Prabhu, 1993). Furthermore,

due to the high solubility of most dextrans in water (above 30 mg/ml), it is very

difficult to remove dextrans from juice by filtration processes (Godshall et al., 1994;

James et al., 2000) and (Yanli.Mi, 2002). On the other hand, the treatment of juice by

the enzyme leads to an increase of dextran low molecular weight fractions, which are

soluble in water, and consequently, reach the crystallization process and cause

several problems in this step.


3

Consequently, the aim of this work was to identify the sources of dextran during beet

sugar production and to characterize the effects of dextran on the crystallization

process and the quality of the final sugar crystal. Also industrially relevant methods

to compensate the effects of dextran in sugar juices during the crystallization process

by variation of the process parameters crystallization temperature and supersaturation

during the crystallization process without any chemical additions should be

identified. In addition, the rate-limiting step in the crystallization process affected by

dextran, elucidation of diffusion and surface reaction phenomena in sucrose solutions

with dextran were studied.

To reach this aim, a specific knowledge is necessary especially about dextran

composition and the influence of dextran on other process variables during the

crystallization process. However, no information was available concerning the effect

of average molecular weight of the dextran fractions on the sugar production process.

OUTLINE OF THE THESIS

The main goals of this study are summarized in the following topics:

Sensitivity and accuracy of different methods for the determination of dextrans of

varying molecular mass.

Identification of microbial sources of dextran and identification of dextran

progression during sugar production.

Characterization of the effects of different molecular fractions and concentration

of dextran on growth rate of sucrose crystals at different crystallization

temperatures. In order to provide representative results in the usual range of

occurring dextrans, molecular mass fractions of M = 40,000 g/mol (T40), M =

500,000 g/mol (T500) and M = 2,000,000 g/mol (T2000) were utilized.

Presentation of an overview about overall crystallization kinetics in order to

explain the relationship between different factors influencing crystallization

kinetics and how they are influenced by dextran themselves.


4

Characterization of the influence of dextran molecular fractions on crystal shape

and surface topography.

Technical and technological consequences.

Review of literature

5

2 Review of literature

Polysaccharides are long chain molecules, either branched or straight. These

molecules are derived from two sources: the metabolic activities of the growing plant

(e.g. starch) and the metabolic activities of microorganisms (e.g. dextran) growing

during its life or at some stage in the subsequent processing (James and Day, 2000;

Wilson, 1996).

2.1 Structure of dextran

A polysaccharide usually referred to as dextran compound widely occurs in

deteriorated sugar cane and beet. These molecules are derived from the metabolic

activities of microorganisms growing during plant cultivation or at some stage in the

subsequent processing (James and Day, 2000).

Figure 1 exemplifies the production of dextran by the action of dextransucrase from

microorganisms, especially by Leuconostoc mesenteroides on sucrose. It can be seen

that sucrose is degraded to a fructose and a glucose molecule. The latter is

polymerized to dextran. Fructose is remaining in solution and can be determined

analytically, a method which is sometimes applied if dextran should be determined

(Clarke and Godshall, 1988).

(Leuconostoc m.)

Dextransucrase

Fructose

Dextran

(Leuconostoc m.)

Dextransucrase

Fructose

Dextran

Figure 1: Dextran formed from sucrose by Leuconostoc m.


6

Dextran is the collective name given to a large class of α-(1→6) linked glucose

polymers (Galea and Inkerman, 1993). Figure 2 shows the chemical structure of

dextran. It illustrates that most linkages in dextran structure are α-(1→6). Dextrans

are from the chemical point of view almost identical to amylose, but they have some

branching of the polymer chain, which prevents this kind of stacking (Walstra,

2003). The linkage in dextran molecules varies from 50 % to 97 % α-(1→6) of total

linkages. The linkages of α-(1→2), α -(1→3) and α- (1→4) are usually at branch

points (Steinbuechel and Rhee, 2005).

Figure 2: Chemical structure of dextrans

The molecular differences in dextrans obviously influence the solubility; for any

given size, the greater the content of α-(1 6) the greater is the solubility.

Conversely, the higher the percentage of α-(1 3) linkages in a polymer the greater is

the decrease in water solubility (Day, 1992).

The degree of branching of dextrans depends on the microbial source and varies

widely among species. For example, while the dextran produced by Betacoccus

arabinosaceous has a unit chain length of only six or seven α-(1 6) linked glucosyl

residues and is highly branched, the dextran produced by Leuconostoc mesenteroides

B-512F may have a unit chain length of greater than 10,000 residues with less than

5 % branching (Kennedy and White, 1988).

An overview to the parameters affecting the particular molecular structure of dextran

is given by (Singleton, 2002). Table 1 shows the different linkage structures found in

dextran. On the other hand, the water soluble, high molecular weight dextran from

Leuconostoc m. NRRC B-512F consists of 95 % α-(1 6) linked α-D-glucopyranosyl

residues with 5 % α-(1 3) linked D-glucosyl or isomaltosyl side chains. However,

n

α (1→6)

α (1→3)

Carbon

Oxygen Hydrogen

Glucose ring

n

α (1→6)

α (1→3)

Carbon Oxygen Hydrogen

Glucose ring


7

Leuconostoc m. NRRC 523 predominately produces a lower molecular weight, water

insoluble dextran which consists of only 66 % (1 6) linkages with 24 % (1 3) and

10 % (1 4) branched linkages (Edye et al., 1995; Jeanes et al., 1954; Robyt, 1986).

Table 1: The linkage structures of dextran

Linkages % Dextran Solubility

α1-6 α 1-3 α 1-3Br α 1-2

L.m.B512F

L.m. B1299

L.m. B1355

S.m. B6715

S.m. B6715

Soluble

Less soluble

Soluble

Soluble

Insoluble

95

66

54

64

4

5

35

94

1

11

36

2

27

L.m. , Leuconostoc mesenteroides; S.m., Streptococcus mutans; Br, Branch linkage.Source: adapted from (Robyt, 1986).

2.2 Microbial loading in sugar factories

Sugar beet is a major agricultural crop in temperate zones of the world, just as sugar

cane is a major crop in tropical zones. Beet harvesting commences when the weather

turns cold enough for storage of the harvested beets. Again, climatic conditions have

a great deal to do with operations. The tops of the beet are removed prior to

harvesting. Generally, harvesting is a mechanical method. A variety of lactic acid

bacteria (LAB), including Leuconostoc species are commonly found on crop plants

(Day, 1992).

Dextran was first reported to be formed from sucrose by strains of Leuconostoc

species in 1930 (Hucker and Pederson, 1930). Later, dextran was obtained in cell-

free extracts from Leuconostoc mesenteroides (Hehre and Sugg, 1942). Different

strains of Leuconostoc mesenteroides produce dextran with different physical,

chemical and serological characteristics (Kobayashi and Matsuda, 1974). The

production of dextran depends on different factors (i.e. type of strain, environmental

conditions and reaction conditions between enzyme and substrate) (UI-Qader et al.,

2001). The yield of dextran increases during growth and maximum yield is obtained

at the end of exponential phase. Dextran yield was higher in media containing

nitrogen source supplemented with different salts (UL-Qader et al., 2005).


8

Morphologically, it is often difficult to separate leuconostocs from streptococci,

lactococci, and lactobacilli, because leuconostocs, lactococci, and streptococci may

form ovoid or even rod-shaped cells (in older cultures or under stress); and

lactobacilli (e. g., Lb. coryniformis or Lb. sake) may produce very short rods or

ellipsoid cells. This explains why, in the past, when identification of leuconostocs

was still based mainly on morphology and slime production, dextran-forming strains

of Lb. confusus were frequently misidentified as L. mesenteroides subsp.

mesenteroides (Holzapfel and Kandler, 1969).

Most strains in liquid culture appear as cocci, occurring single or in pairs and short

chains, however, morphology can vary with growth conditions. Cells grown in

glucose or on solid media may have an elongated or rod shaped morphology. Cells

are Gram positive, asporogenous and non-motile (Figure 3) (SCIMAT, 2006).

Leuconostoc species are differentiated from other lactic acid bacteria on the basis of

phenotypical criteria such as coccoid appearance, formation of gas, inability of

arginine hydrolysis, and production of the D (–)-lactate isomer from glucose, but the

formation of dextran from sucrose and cell wall composition may also be helpful for

identification (Holzapfel and Kandler, 1969).

The sugar fermentation pattern is of limited value for identification of the non-

acidophilic leuconostocs because of the considerable variation among different

strains of the same species on the one hand and because of the similarity of the range

of fermented sugars of different species on the other (Holzapfel and Schillinger,

1992).


9

Leuconostoc citreumBar: 1 µm

Leuconostoc cremorisBar: 2 µm

Leuconostoc lactisBar: 1 µm

Leuconostoc mesenteroidesBar: 1 µm

Figure 3: Micrographs of four strains of Leuconostoc spp by scanning electron microscopy (SCIMAT, 2006)

Out of 15 different sugars, 13 may be fermented by strains of Leuconostoc

mesenteroides subsp. dextranicum and there are also many differences in the

fermentation pattern among different strains of Leuconostoc mesenteroides subsp.

mesenteroides. Only Lc. mesenteroides subsp. cremoris is easily distinguished from

the other leuconostocs by its very limited range of fermented sugars, consisting of

glucose, galactose, and lactose. Sugars most helpful for differentiation of the other

species seem to be arabinose, melibiose, trehalose, and xylose (Holzapfel and

Schillinger, 1992).

Hamasaki et al., (2003) reported that the optimum growth temperature of L.

mesenteroides subsp. mesenteroides was 30°C and the maximum was 37°C (Figure

4).

A recent study of Guglielmone et al., (2000) showed that Leuconostoc mesenteroides

(strain 3A) consumes sucrose very quickly (8.05 g/l/hr at 25 °C and 8.46 g/l/hr at

30 °C) during the first 6 hours of culture. This fermentative process implies a sucrose

consumption of 59 % at 25 °C and 62 % at 30 °C. At higher temperatures (37 °C and

40 °C) the percentage of consumed sucrose decreases to 47 % and 27 % respectively.


10

Incubation Time (hours)

0 6 12 18 24

AB

S (6

60 n

m)

0.0

0.5

1.0

1.5

Figure 4: Growth curves of Leuconostoc mesenteroides subsp. mesenteroides MCRI 1 in MRS broth at 30-40°C (Hamasaki et al., 2003) (30°C (•), 35°C ( ), 37°C ( ), and 40°C (X). ABS,

absorbance)

The strain of organism studied was isolated from sugar cane juice in Argentina. Its

high fermentation rate consumed sucrose rapidly, stopping its growth and sugar

utilization 6 hours after incubation. From the sugar industry point of view, it is

important to know the consumption of sucrose in short periods as shown in (Table 2).

This is due to the importance of dextran in the sucrose loss by the direct sucrose

consumption as a source of energy (see Figure 1) and production of dextran, which

causes a reducing in the efficiency of manufacturing processes.

Table 2: Sucrose consumption and dextran production at different time intervals, Leuconostoc mesenteroides 3A; culture medium 10% sucrose basal medium; incubation temperature 30 °C

(Guglielmone et al., 2000)

Time (hours)

Consumed sucrose (g/l)

Dextran production (g/l)

1 08.50 0.35 2 17.00 0.60 3 25.40 0.80 4 33.80 0.90 5 42.30 1.08 6 50.80 1.30 9 51.90 2.25

15 53.40 5.10 18 54.10 7.00 24 55.50 13.00 48 57.50 13.20


11

2.3 The common methods of dextran fractions determination

In the sugar industry, several different methods are in use for the determination of

dextran. These methods are Haze method, ASI II method, Robert’s copper method,

Midland SucroTestTM, Optical Activity Ltd. DASA method and HPLC method

Nicholson and Horsley, (1959) developed the forerunner to the current haze assay.

This procedure involves the separation of material precipitated with alcohol to make

a 70 % v/v solution. The precipitated material is resolved in water and treated with α-

amylase to break down starch and trichloracetic acid (TCA) to precipitate proteins.

The dextran haze is quantified in a spectrophotometer at 720 nm. The absorbance is

compared with that obtained on a standard curve using dextran isolated from

Leuconostoc mesenteriodes (Day et al., 2002; Paton et al., 1994).

After degradation of starch by amylase, the ASI II method treats alcohol-precipitated

samples with dextranase and α-glucosidase. This results in the conversion of dextran

to glucose. The glucose produced is then quantified by the Nelson-Somogyi

arsenomolybdate method (Sarkar and Day, 1986).

Sarkar et al., (1991) compared the Haze test with the above method (ASI II method),

a GPC method (Saska and Oubrahim, 1987) and a polyclonal and monoclonal

antibody test. The ASI II method, being an enzyme based method, was suspected to

have problems at low dextran concentrations due to high degrees of branching. The

GPC method was difficult to quantify and not sensitive to all molecular weights. The

antibody tests did not detect dextran with a molecular weight below 150 kDa.

Another quantitative method for dextran detection has been proposed by Roberts

(Roberts, 1983) to compete with the haze assay. Like the original Nicholson and

Horsley method, all polysaccharide and protein constituents are precipitated and

redissolved in water. Then dextran is precipitated ”selectively” with copper sulphate.

The dextran from this latter precipitation is determined colorimetrically by phenol-

sulphoric acid. It was claimed that the Roberts method was independent of the

molecular weight of the dextran and was specific for dextran (Paton et al., 1994).

Sancho et al., (1998) devised a test based on stripping voltametry, which showed a

linear relationship to dextran concentration in standard sucrose solutions. The

performance of the procedure is compared with the Robert test (Roberts, 1983) and


12

performed better, especially at low dextran concentrations, but still in standard

solutions. The method is claimed to be faster and simpler than the Roberts test, but it

needs further investigation using real sugar samples.

Brown and Inkerman, (1992) developed an HPLC method that uses dextranase to

break the dextran down into isomaltose, which is subsequently quantified by HPLC.

The method was proposed as a reference method (Urquhart et al., 1993). However,

the procedure is technically difficult and time consuming. The dextran is isolated

using 80 % ethanol and then hydrolyzed with dextranase. The reaction is completed

and the amount of isomaltose produced is converted into dextran concentration by a

calibration function which bases on the usage of dextranase on standard dextrans.

The method has been critisised for assuming a homogeneous dextran structure.

Two new methods, the Midland SucroTestTM and the Optical Activity Ltd. DASA

methods are reported to determine only dextran. Midland SucroTestTM is an

immunological method that uses a monoclonal antibody to produce a linear increase

in turbidity with response to dextran. The dextran concentration is determined from

the change in turbidity after addition of the sample into a buffer solution containing a

monoclonal antibody and comparing the difference to a standard. Midland

SucroTestTM showed consistent results and a high correlation with the Haze test

(Day et al., 2002). This assay method is a rapid test that utilizes a minimum of

equipment, and is highly specific to dextran, but the cost per analysis is considered

too high for extensive use.

The DASA method uses a near infrared spectrometer to detect the change in optical

rotation in a sample prior to and after dextranase treatment. The dextran

concentration is calculated and displayed as a change in optical rotation in an OA

Ltd. SacchAAr 880 polarimeter. The DASA method was reported to give

erroneously high levels of dextran, due to either partial dextran hydrolysis or poor

filtration (Singleton et al., 2001).

The new monoclonal antibody test method is more sensitive and allows testing of

various process streams. Of great importance is, that the test can be conducted in

only three to five minutes, which is a significant improvement compared to all other

methods. The antibody attaches to dextran and produces a haze. The amount of haze


13

formed is directly proportional to the amount of dextran present. A specially

modified nephelometer (turbidimeter) is used to measure the turbidity of the haze

formed by the reaction of the antibody with dextran. However, this method is very

expensive and not available for use in sugar factories (Rauh et al., 1999).

2.4 Dextran content during the process of sugar production

Before dealing with the progression of dextran a short introduction to the sugar

production process should be given (comp. Figure 5). After harvesting, the beets are

hauled to the factory. Each load entering is weighed and sampled before tipping onto

the reception area, where it is moved into large heaps. The climatic and storage

conditions play an important role regarding the deterioration of beets (Kenter and

Hoffmann, 2006). The beets are conveyed into the factory in a water flume. Rocks

and weeds are removed prior to the beets being washed. The washing process

contributes to a reduction of dextran levels in the extracted juice. The washed beets

are then sliced into thin noodle-like strips called cossettes, which are then conveyed

to the extraction system. The aim of the diffusion is to extract the maximum amount

of sucrose with the minimum of impurities (non-sugar substances). The dry

substance and pH of extracted juice range from 10 to 18 % and 5.4 to 5.6,

respectively, which are quite favourable conditions for multiplication of

microorganisms such as Leuconostoc, Streptococcus and Lactobacillus spp., etc. , the

clarification process of juice practically eliminates the microorganisms present in it

(Gupta and Prabhu, 1993). To limit the thermophilic bacterial action, the feed water

may be dosed with formaldehyde. A control of feed water pH is also practiced (van

der Poel et al., 1998).

During juice purification, the juice from the extraction process is separated from

other non-sugar substances. Therefore, the juice is progressively heated and lime is

added at three different points in the juice purification system. The high temperature

and the lime break down the main part of the impurities in solution. Carbon dioxide

is then added and calcium carbonate precipitate is formed. The juice-calcium

carbonate mixture is then filtered to remove the calcium carbonate leaving the clear

juice. Dextran formation occurs during storage and the extraction process, whereas, it

is reduced during the clarification and filtration processes (Gupta and Prabhu, 1993).


14

The juice must then be concentrated by a multi-stage evaporator. The resulting thick

juice contains about 70 % solid. Some part of the thick juice is stored in large outside

tanks for processing after all harvested beets are processed. This allows the factory to

slice about 50 % more beets than the crystallization capacity can handle. Sugar is

crystallized from the thick juice by continued evaporation of water in vacuum pans.

The crystals are separated from the syrup in centrifugal machines. The crystals are

washed with water and then dried in a rotary drum drier (McGinnis, 1982; van der

Poel et al., 1998).

Looking specifically at the occurrence and progression of dextran in sugar

production, most of the available examinations are reported from trials in sugar cane

plants. Usually, dextran levels in sugar cane processing are higher than in sugar beet

processing as sugar cane grows in warm areas and is more vulnerable to infection by

microorganisms. Only few information is available about the progression of dextran

in sugar beet factories. In the following paragraph a short overview of reports should

be given.

Dextran producing bacteria are found in high sucrose areas, including the soil in

sugar cane fields, in sugar juices and on exposed equipment in cane and beet sugar

mills and sugar refineries (Clarke and Godshall, 1988; Robyt, 1986). Dextran levels

in cane are usually controlled by good planning of cane deliveries, and good hygiene

in the cane yard, the mills and the factory. However, there are times when weather

problems (storms, freeze) cause unavoidable damage to cane and delay deliveries. In

these cases, infection and dextran levels build up in cane before it reaches the factory

(Atkins and McCowage, 1984; Inkerman, 1980).


15

Extracting the juice Purification of the juice

Evaporation of the juice Crystallization

Figure 5: The major steps for the manufacture of sugar from sugar beets (Südzucker AG website, Germany)


16

Dextran levels as high as 10,000 mg/kg DS have been reported in syrup (Wells and

James, 1976). The level of dextran in raw sugar is of commercial importance because

it affects the purchase price for raw sugar (Clarke and Godshall, 1988). Levels in

raw sugar range up to 5000 mg/kg (total dextran) and in refined sugar up to 2500

mg/kg (total dextran). They added, that in recent years dextran concentration in raw

sugar has been risen because of the increased use of mechanical harvesting systems

which tend to damage the cane, allowing entry of the causative organisms into the

cane stalk (Khaleifah, 2001). As an example Table 3 documents the high dextran

levels in different Louisiana sugar factories (Cuddihy and Day, 1999).

Table 3: Dextran in process streams measured at different Louisiana sugar factories(F1, F2, F3 and F4) (Cuddihy and Day, 1999)

Dextran mg/kg DS* Samples

F1 F2 F3 F4

Mixed Juice 2690 1094 1928 4650

Clarified Juice 2181 1094 1928 4560

Syrup 2602 1239 1986 3858

* Dextran measured by ASI2 method according to (Sarkar and Day, 1986)

2.5 Dextrans associated with processing problems

In sugar production, dextrans are undesirable compounds produced by contaminant

microorganisms from sucrose (Jimenez, 2005). On contrast, dextrans are used in the

manufacturing of blood plasma extenders, heparin substitutes for anticoagulant

therapy, cosmetics, and other products (Alsop, 1983; Kim and Day, 2004; Leathers et

al., 1995; Sutherland, 1996).

From an industrial point of view, studies have shown that Leuconostoc

mesenteroides strains are able to utilize high percentages of the sugar present in

juices in a short time period. This implies important losses if the infection level is not

controlled. High loss in product yield (sucrose) and contamination of the industrial


17

process cause various problems such as: an increase of juice viscosity which

produces blockage in the process line, pumps and filters; lower heat exchange;

evaporation diminution; decrease in the efficiency and output of crystallization;

crystal shape distortion; blockages in centrifuges and sucrose losses to molasses.

Dextran in juice, syrup and sugars can cause false polarization. Because dextran is

highly dextrorotatory, approximately three times that of sucrose and gives a falsely

high polarization, unless removed prior to test. Furthermore, high dextran levels

reduce the efficiency of clarification techniques used in polarization determinations

(Clarke et al., 1997; Cuddihy et al., 2000; Guglielmone et al., 2000).

In addition, the farmer is mainly paid based on the polarimeter reading, therefore,

there is an obvious need for a dextran test in the core laboratory. The benefits arising

from this would be the correction of the falsified reading and the identification of the

sources of dextran contamination entering the factory (Singleton et al., 2001).

Measurements of the sucrose content of juices by NIR polarimetry, at 880 nm, are

not affected by the yellow/brown color remaining after conventional filtration, using

a filter aid. Readings obtained using NIR polarimetry in comparison to those at the

sodium wavelength have been previously shown to be more reproducible, and more

sensitive to interference by high dextran concentrations (Wilson, 1996).

The NIR polarimeter is based on a linear relationship between the change in optical

activity produced by enzymatic cleavage of dextran and the original dextran

concentration. There is no significant difference in the initial change of optical

activity produced by enzymatic hydrolysis of different molecular weight dextrans

using dextranase (Singleton, 2002).

At the clarification stage in sugar production, abnormally high dextran levels cause

severe effects. Juice derived from deteriorated cane contains excess acids and

requires extra lime addition to neutralize the acidity. The presence of excess gums

increases juice viscosity, which retards the setting time in clarifiers and prevents

satisfactory removal of suspended matter. Consequently, the clarified juice is cloudy,

resulting in higher mud volumes at the filter station. Filtration of mud is also

impeded by the excess viscosity of the juice. Furthermore, Jimenez, (2005) reported

that the high viscosity of the juices and the presence of high molecular weight


18

dextrans in them along with other insoluble matter cause a blockage of the filters

provoking juice losses due to spills that are generally underestimated.

The changes in the physical properties of sugar solution have been attributed to

dextran mediated viscosity alterations. Effects are changes in heat transfer

characteristics and changes in the crystallization characteristic of the sucrose. The

most damaging effects of high levels of dextran are seen on crystal growth.

Laboratory studies in Australia have shown a sugar loss to molasses of between 1.2

to 1.4 purity points, which can be, expected for every 1,000 mg/kg DS of dextran in

molasses. This is equivalent to about 250 mg/kg DS in mixed juice (Miller and

Wright, 1977).

The problems associated with dextran contamination, in both the factory and the

refinery, are well documented in the literature and thus are briefly summarized in

Table 4.

Table 4: Summary of the detrimental effects of dextran in terms of the losses it leads to according to Singleton et al., (2001).

Production losses Sucrose losses Direct financial losses

Increased viscosity leads to

reduced throughput due to:

Poor filtration rate

Reduced evaporation rate

Reduced flocculation rate

Slow mud settling

Poor crystallization

(elongation)

Dextran

formation

To molasses

(melassigenic

effect)

False polarimeter reading

leads to overpayment to

farmer

In trade of raw sugar as

part of dextran penalty

system using unreliable

tests


19

2.6 Crystallization process

The most damaging effects of elevated dextran concentrations in a technical sucrose

solution are foreseen in the crystallization process. Dextrans slow down the

crystallization rate or even inhibit crystallization (i.e., they have a high melassigenic

effect).

There are different methods for sucrose crystallization and to follow the growth rate

of sucrose crystals during crystallization process. In this work, the isothermal

crystallization method was used to study the effect of dextran molecules on the

crystallization process without interference of other factors. Different authors in

literature focused the characterization of parameters influencing the crystallization

process as well as mathematical approaches for the description of crystallization

during sugar production.

2.6.1 Growth rate of sucrose crystals

The sucrose crystallization process is the transfer of sucrose from the solution phase

to the solid phase. Supersaturation is the measure of the driving force to cause this

transfer to take place.

There are three simple basic methods to grow crystals from a solution

The evaporation method

The slowly cooling method

Isothermal crystallization method

The evaporation and the cooling method require the beginning with a saturated sugar

solution but the isothermal crystallization method requires beginning with a

supersaturated sugar solution. The evaporation method employs the use of heat to

separate the water from the sugar. The slow cooling method produces sugar crystals

by letting a hot saturated sugar solution cool down slowly. The slower the process,

the bigger sugar crystals are formed. The process may take several hours to several

days. The isothermal crystallization method as a laboratory method uses a constant

temperature for crystallization. The crystallization rate can be calculated by

following the progression of dry substance (WDS) in solution during the

crystallization process.


20

Six general systems for initiating crystals are recognized by McGinnis, (1982):

1- Homogeneous – new crystal particle formation as the result of

supersaturation only.

2- Heterogeneous - new crystal particle formation as the result of

supersaturation and the presence of foreign insoluble material.

3- Secondary - new crystal particle formation as the result of supersaturation and

in the presence of an ongoing, growing crystal crop.

4- Attrition – crystal population increase due to the fracture ore impact chipping

or breaking of existing, growing crystals into smaller particles, each capable

of growth.

5- Full Seeding – preparation of seed crystals by salting-out, grinding,

screening, or centrifugal separation, mixing as a slurry or magma; adding the

mixture as a controlled population into supersaturated liquor for growth only.

6- Ultrasonic – liquid conditions similar to homogeneous except a much lower

supersaturation. Ultrasonic irradiation creates mechanical perturbations

adequate to exceed the energy barrier of homogeneous nucleation.

According to Gillet, (1977) and Elahi, (2004), the success of the crystallization

process depends on several factors as follows:

Supersaturation of solution

Crystallization temperature

Relative velocity between crystals and mother liquor

Nature and concentration of impurities in solution

The crystal surface area

Schliephake and Ekelhof, (1983) and Ekelhof, (1997) explained the relation between

the growth rate of sucrose crystals and many parameters such as supersaturation,

temperature and relative velocity between crystals and solution during the

crystallization process. The relationship between the growth rate of sucrose crystals

and the supersaturation of solutions with different purities is shown in Figure 6. The

increase of solution purity leads to an increase of growth rate.


21

q = 100%

q = 90%

q = 80%

q = 70%

q = 60%

1.0 1.1 1.2 1.3 1.4

Supersaturation coeffic ient

Cry

stal

gro

wth

in g

/(m2.

min

)

1 .2

0.9

0.6

0.3

0

Figure 6: Growth rate at 60 °C as a function of supersaturation at different purities, (Schliephake and Ekelhof, 1983)

Another important factor affecting the rate of sucrose crystallization is temperature.

Figure 7 shows the strong temperature effect on the crystal growth rate, a factor of

particular importance in cooling crystallization.

In addition Pot et al., (1984) investigated the effect of crystal size on the crystal

growth rate and found that the growth of crystals >100 µm is determined by the

diffusion velocity.

0 20 40 60 80

Temperature in °C

q = 100%

q = 90%

q = 80%

q = 70%

q = 60%

1.2

0.9

0.6

0.3

0

Cry

stal

gro

wth

in g

/(m2.

min

)

Figure 7: Crystal growth rate as a function of temperature at different purities and constant supersaturation (Schliephake and Ekelhof, 1983)


22

The crystal growth rate is also affected by high molecular mass substances and

colorants in syrups, which are partially adsorbed on the crystal surface and thus

impede the insertion of sucrose molecules into the lattice (Grimsey and Herrington,

1996; Rogé et al., 2007).

2.6.2 Crystallization kinetics

Two fundamental crystallization theories are briefly described: the theory of the

adsorption layer, which deals with the crystal surface; and the diffusion theory,

which mainly considers the phenomena occurring at the crystal solution interface.

Another approach is proposed based on the chaos and complexity theory.

The chaos and complexity theory deals with the nonreversible thermodynamics of

the conditions apart from equilibrium. It is correlated also with the concept of

entropy, which, being a measure of disorder and uncertainty becomes the cause for

creation of new patterns of mass and energy dissipation. These are called “dissipative

structures from the 1977 Nobel laureate Ilya Prigogine”. Apart from the different

levels of concentration, viscosity, temperature, pressure, supersaturation, etc., of

mother liquor, there are different impurities, which influence sucrose crystal growth

morphology, modifying the habit of the crystals. The sugar crystal is an “equilibrium

structure but bears on its surface and on its inner structure as fingerprints the history

of a dissipative process” (Cbristodoulou, 2000).

The relation between the diffusion, surface reaction and crystallization rate of

sucrose was illustrated by (Cossairt, 1982; Ekelhof, 1997; Houghton et al., 1998;

Silin, 1958). If the diffusion theory is considered, the variation of the solution

concentration near the surface hast to be regarded.

The crystallization rate is given by a diffusion process of molecules through the

stagnant layer and the reaction process corresponding to the integration of molecules

into the crystal structure through the adsorption layer. These two steps can be

illustrated by equations 2-1 and 2-2:

[ ]GLD ccAkdtdm

−⋅⋅= (Diffusion) (2-1)

[ ]SatGR ccAkdtdm

−⋅⋅= (Surface reaction) (2-2)


23

Where

dtdm / Crystallization rate (were m mass and t time)

D Diffusion coefficient (m2/s)

Dk Coefficient speed of the material transfer (m/s)

Rk Coefficient speed of surface reaction

A Crystal surface (m2)

Lc Concentration of solution (kg/m3)

Gc Concentration of boundary layer (kg/m3)

satc Mother liquor concentration at saturation point (kg/m3)

The Noyes and Whitney’s formula (Noyes and Whitney, 1897) below for

crystallization is used to review the basic control parameters of the crystallization

process:

( )1−⋅⋅⋅= SSCAkdtdm

satd (2-3)

where = Csat = concentration of saturated liquor and SS = supersaturation

Schliephake and Ekelhof, (1983) reported that, the change of the supersaturation at

the same time, the difference of concentration between concentration of solution (CL)

and concentration of saturated solution (Csat) causes a change of the crystal growth

rate. Thus, they derived the following equation to calculate kD from the above

equations:

( ) ⎟⎠⎞

⎜⎝⎛ ⋅−Δ−Δ

⋅⎟⎠⎞

⎜⎝⎛ ⋅

=

Adtdmcck

kAdt

dm

k1*

1

R

R

D (2-4)

where ( )cc *Δ−Δ is the effective concentration difference (kg/m3)

The parameters Dk and Rk can vary depending upon the crystallization conditions,

and in particular upon temperature, stirring and the presence of non-sugars

(Mantovani and Vaccari, 1998).


24

There are many parameters which have an effect on the diffusion coefficient kD.

(Schliephake and Ekelhof, 1983) described the coefficient as

( )εηρ ,,,,, LLLKLD DdTfk = (2-5)

where

TL Temperature of solution

dK Crystal diameter

ρL Solution density

ηL Solution viscosity

ε Relative velocity volume

The relative velocity volume was calculated by the following equation:

1

1

−+

−

=

Cry

L

Cry

Su

Cry

Su

mm

mm

ρρ

ε (2-6)

where

Sum Crystal suspension mass

Crym Crystal mass

Cryρ Crystal density

From the above equations, it can be stated that not only temperature can effect the

diffusion coefficient but also crystal diameter, density of solution, relative velocity

and solution viscosity which plays an important role in the diffusion processes in the

solution. Also, it can be indicated that the surface reaction coefficient is affected by

temperature and solution purity, where the other factors are constant

The surface reaction rate kR was calculated by means of the equation 2-7 for any

temperature.

RTE

0,

Ra,−⋅= ekk RR (2-7)

where Ea = activation energy of the surface reactions = 86.2 kJ/mol, R = general gas

constant = 8.3147 J/mol*K, T = absolute temperature (Kelvin) ∞k = Frequency

factor. For pure sucrose solutions can be kR can be calculated with the help of

equation 2-7 for any temperature as follows:


25

)2.86(7182818.2*8892.3 RT

R Ek−

+= (2-8)

As far as temperature is concerned, it is well known that, at low temperature, the

crystallization process is controlled by the surface reaction whereas at high

temperature it is diffusion, which controls the crystallization process. These results

are similar to those reported by (Cossairt, 1982; Houghton et al., 1998).

The correlation between temperature and the surface reaction rate kR was illustrated

by (Ekelhof, 1997). The temperature dependence of the determined kR-values is a

linear connection like expected from Arrhenius relation (Figure 8) (Golovin and

Grerasimenko, 1959; Maurandi et al., 1988; Schliephake and Austmeyer, 1976) .

Figure 8: Surface reaction coefficient kR dependent upon the reciprocal value of the absolute temperature T given for different purities (Ekelhof, 1997).

As expected by the recently introduced “spiral nucleation model”, this alternative

definition is found to be size-independent over the considered supersaturation range.

At the same time, the conventional overall growth rate expressed per time and

surface area units is found to be linearly dependent on crystal size. Besides being

theoretically consistent, the volumetric growth rate concept is of great practical


26

interest since crystal growth kinetics can be calculated in situations of unknown

crystal number and size. The two-way effect of crystal size on mass transfer rates and

on the integration kinetics is investigated by measuring the sucrose dissolution rates

under reciprocal conditions of the growth experiments. Both effects are adequately

described by combining a well-established diffusion-integration model and the spiral

nucleation mechanism (Martins and Rocha, 2000).

For a deeper insight into the theoretical growth morphology it is necessary to apply

the HP theory of Hartman and Perdok, (1955) on the crystal structure. This theory is

based on the study of the periodic bond chains (PBCs) that form during

crystallization. The faces are classified into three types, in the first case the faces are

so called F-faces (Flat), in the second case S-faces (Stepped) and in the third case K-

faces (Kinked). They occur as shown in Figure 9. The three types of faces have

different growth behaviors depending upon the different density of the growth sites

(kinks).

Figure 9: Hypothetical crystal (a) with schematic drawing of the three kinds of faces PBCs (b)

The several steps in the insertion of a molecule from solution into the crystal lattice

are shown in Figure 10. The driving force in this process is the supersaturation

maintained by either water evaporation or temperature reduction. With increasing

amount of evaporated water during evaporation crystallization, the crystal surface

becomes inadequate, the supersaturation increases and so a new crystal surface is

formed by false grain formation (Schiweck and Mannheim, 1998).


27

1

23

6

77 + 8

94

7

7

5

Step

Crystal half layer

1

23

6

77 + 8

94

7

7

5

Step

Crystal half layer

Figure 10: Diagrammatic representation of the processes involved in crystal growth from solution, according to Elwell and Scheel ,(1975) as modified by Kruse and Ulrich, (1993).

The description of Figure 10 can be summed up as follows: 1 Diffusion of hydrated

crystal building blocks from the solution to the crystal surface. 2 Adsorption of

hydrated crystal building blocks (cluster) and partial dehydration on the crystal

surface. 3 Surface diffusion of hydrated crystal building blocks to a step. 4

Adsorption of hydrated crystal building blocks and partial dehydration on the step. 5

Diffusion of hydrated crystal building blocks along a step to a semi-crystal position.

6 Insertion of crystal building blocks into a semi-crystal position and complete

dehydration; 7 Diffusion of the liberated hydrate envelope into solution. 8 Liberation

of latent heat on insertion of crystal building blocks into a semi-crystal position. 9

Desorption of hydrated crystal building blocks from the crystal surface into the

solution.

2.6.3 Parameters influencing crystallization kinetics

Whatever the origin, the polysaccharides from beet processing are found as minor

constituents in white sugar and their association with nitrogen and cations can cause

the coloration of the crystals. These polysaccharides generally have a high

hygroscopicity and their presence at the surface of sugar crystals my affect the

stability of crystalline sugar during storage and provoke caking. Moreover, occlusion

and crystal elongation (c-axis) are observed in presence of dextran (Gudshall, 1992;

Parrisch and Clarke, 1983).

During processing, macromolecules (e.g. dextran) increase viscosity, slow or inhibit

crystallization, and increase the loss of sucrose to molasses (i.e. they have a high

melassigenic effect). Because of their carbohydrate nature and high solubility, they

are difficult to remove in processing, and tend to be included in the raw sugar crystal,


28

going into the refining process and causing similar problems in refining (Cuddihy et

al., 2000; Godshall et al., 1994).

The viscosity is a dominating factor in the technology of the process. The growth rate

is sensitive to variation in crystal size and slightly influenced by surface integration,

with the diffusion step in crystal growth as a significant factor in determining the

overall rates (Gumaraes et al., 1995). Frenkel, (1958) has found that the absolute

viscosity of sugar solutions is related to the absolute temperature by an equation of

the form:

TBA /10*=η (2-9)

where:

η Dynamic viscosity

A and B Constants

T Temperature

Figure 11 shows the relationship between the viscosity in pure and impure solution

and solution concentration at different temperatures (40 to 80°C) according to

Schliephake and Ekelhof, (1983). They illustrated that the solution viscosity

increased with a decrease of temperatures. However, it increased with the increase of

solution concentration. These correlations between relative viscosity ru ηη / and other

factors were described as follows:

( ) ⎥⎦

⎤⎢⎣

⎡−

⋅−−

−=DS

DS

r

u

wwqn1819

0379.0734.44.0

11 ϑη

(2-10)


29

0 2 0 4 0 6 0 8 0

W D S %

1 0 0 0

1 0 0

1 0

1

ηm

Pa.s

Figure 11: The viscosity of the pure and impure sucrose solution according to Ekelhof, (1997)

2.6.4 Crystal morphology

The normal morphology shown by sucrose crystals grown from pure solution is

depicted in Figure 12 (Vavrinecz, 1965). The three primary hydroxyl groups at the C-

1' and C-6' (Fructose) and the C-6 (glucose) belong to the three hydroxymethyl side's

groups of the sucrose. The others are secondary hydroxyl groups at the heterocyclic

ring. The average values of the surface angles for proportional axis were measured

by different investigators as, a : b : c = 1.2543 : 1 : 0.8878 with an angle of the a to

the c axis of β = 103°30 (Vavrinecz, 1965). So far 15 different surfaces were

observed, which arise most frequently with the technical sucrose crystallization.

Under the influences of non-sugar materials and other factors, which affect on the

growth conditions such as e.g. temperature and supersaturation of solution, the

surface growth and thus the habit of the sucrose crystal can vary strongly (Bubnik et

al., 1992; Vaccari et al., 1999; Vaccari et al., 2002). The presence of

oligosaccharides such as raffinose, which is present in sugar beets, causes an

elongation of the crystal toward the b-axis to a needle form (Mantovani et al., 1967;

Vaccari et al., 1986). Dextran also causes an elongation of the crystal toward the c

axis due to slowing down the growth of the p and p’ surfaces (Sutherland and Paton,

1969).


30

Figure 12: Schematic representation of a sucrose crystal with the surface designation according to Vavrinecz ( 1965)

Bubnik et al., (1992) studied the relationship between the c/b ratio if dextran

concentration was increased (Figure 13). They observed that the effect of dextran on

crystal elongation was much more striking compared to that obtained with invert

sugar and 300 % higher than for pure solution, at the highest impurity concentration.

This correlation between c/b and dextran concentration was described as follows:

7.0do

1.0dosd cbcaRR ⋅+⋅+= (2-11)

where:

Rd c/b ratio in the presence of dextran

Rs c/b ratio for pure sucrose solution (= 0.70)

cd Dextran concentration (g/100 g water)

ao Constant (0.1363)

bo Constant (0.1234)


31

+C

+C

+b+b

(A)(A) (A)(A)(A)(A)

Figure 13: (A) Sucrose crystal growth in the presence of dextran (40g/100g water), (B) One of the crystals shown in (A) and (C) Sucrose crystal growth in the presence of glucose, fructose and

dextran (50g/100g water each) (According to Bubnik et al., (1992)

Also, it was possible to calculate a two-parameter function which would permit

calculation of the c/b ratio as a function of both invert sugar and dextran

concentrations:

7.01.0ddi cBcARR ⋅+⋅+= (2-12)

R c/b ratio in the presence of impurities

Ri c/b ratio in the presence of invert sugar

cd Dextran concentration (g/100g water)

with 7.0

21 iio cacaaA ⋅+++=

ao Constant (0.1363)

a1 Constant (0.102)

a2 Constant (-0.00193)

c Invert sugar concentration

with Ciio cbbB *+=

bo Constant (0.1234)

b1 Constant (-0.0147)

C Constant (0.270)

ci Invert sugar concentration


32

Dextran products of various molecular weight result from infections of Leuconostoc

sp. Also, a needle like elongation of the c axis of the crystal may be caused by highly

1-6 linked dextran. However, 1-4 and 1-3 linkages also yield refractory and slow-

boiling syrups, but show little c axis elongation (Cossairt, 1982). It is also reported

that smaller molecular weight dextrans are especially involved in the effect of c axis

elongation (Singleton, 2002)

From the processing point of view the development of re-entrant angles with their

effects is becoming worse as aggregates become more complex. These re-entrant

angles are traps for mother liquor and zones, which are extraordinarily difficult to

wash for the removal of impurities (non-sugars). The trapped mother liquor may

even be present as inclusions between the individual conglomerated crystals

(Houghton, 1998).

The effect that dextran has on sugar crystal shape, to elongate the crystal or cause

needle grain, also increases the loss to molasses by blinding centrifugal screens with

elongated crystals (Imrie and Tilbury, 1972)

2.7 The Economic gain

In a sugar factory, the crystallization is an important step, which determines the yield

of the sugar from sugar beet and cane. About 85% of the present sugar (sucrose) in

sugar beet can be crystallized as sugar product (Figure 14). The sugar losses are

almost 3% during the extraction and clarification processes and around 12% to

molasses (Ekelhof, 1997).

The existence of quality points for dextran in raw sugar contracts has provided a

standard for a penalty by refiners. The penalty is calculated from a sliding scale

based upon the sugar dextran concentration.


33

Purification

Evaporation

Multi-Stages for Crystallization 85%

100% S

3%

12%

White Sugar

Extraction

Beet

Molasses

Purification

Evaporation

Multi-Stages for Crystallization 85%

100% S

3%

12%

White Sugar

Extraction

Beet

Molasses

NS

Figure 14: Partitions of sugar loss during different process stages of sugar production (Ekelhof, 1997)

Studies in Australia have shown a sugar loss to molasses of 1.2 to 1.4 purity points,

which can be expected for every 1,000 mg/kg DS of dextran in molasses. This is

equivalent to about 250 mg/kg DS in mixed juice (Miller and Wright, 1977).

It is estimated that for every 300 mg/kg DS dextran in syrup there is a 1% increase in

molasses purity (the percentage ratio of sucrose in total solids in a sugar solution)

(Atkins and McCowage, 1984; Clarke et al., 1997; Godshall et al., 1994;

Guglielmone et al., 2000).

The presence of dextran above 250 mg/kg DS on solids in raw sugar brings forth

penalties from the sugar refiner (Chung, 2000; Day, 1992). Effects of dextran in juice

are shown in Table 5.


34

Table 5: Effects of dextran in juice on molasses purity

Dextran in juice (ppm/WDS)

Effect

0

250

500

1000

1500

3000

None

1 point loss in molasses purity

2 point loss in molasses purity

Detectable crystal elongation, 5 point loss in molasses purity

Significant operational problems

Severe problems

Source: Adapted from Chung, 2000; Day, (1986).

The economic losses caused by dextran are continuous throughout the process, since

its early content in the juices falsely increases the amount of sugar calculated for

them and alters the production indicators of the factory. This is due to the

dextrorotatory characteristic of dextrans that polarize approximately three times more

than sucrose producing a high false polarization value (Jimenez, 2005).

The evaluation of dextran levels in raw sugars with the respective sucrose losses

incurred to produce these dextran levels and the amounts of fructose and acids (about

30% yield) formed is shown in Table 6

Table 6: Dextran levels and sucrose loss in raw sugars (Bose and Singh, 1981)

Dextran

(%)

Sucrose loss

(Kg /ton sugar)

Fructose formed

(Kg /ton sugar)

% Acid produced

0.05 0.20 0.99 0.07

0.10 0.40 1.98 0.17

1.50 2.00 9.90 0.70


35

These amounts of sucrose loss represent only those lost directly due to dextran

formation. Three to five times of these levels may be lost later in processing because

of the other organic material formed conjointly with dextran (Bose and Singh, 1981).

Dextran content in sugar is a major concern for end users such as candy

manufacturers. Contamination of the sugar with dextran, above a certain ppm level,

will affect hard candy processing. The impact is measured in changes in candy

thickness/weight and is related to dextran content in sugar (Haynes, 2004; Jimenez,

2005).

Material and methods

36

3 Material and methods 3.1 Material

Samples: sugar beet slices (cosettes), raw juice, clarified juice, thick juice and the

final products, sugar and molasses were collected from three different Egyptian

factories. Factories 1 and 2 belong to Delta Company in north Egypt and factory 3

belongs to Egyptian Sugar Company in middle Egypt.

Dextrans: Three different molecular mass fractions of dextran (T40: M~40,000

g/mol, product Nr. 31389, T500: M~500,000 g/mol, product Nr. 31392 and T2000:

M~2,000,000 g/mol, product Nr. 95771) from Sigma-Aldrich were utilized. All

fractions were produced by Leuconostoc ssp.

Chemical for analysis: all chemicals were obtained from Sigma-Aldrich, Germany.

3.2 Analytical methods

3.2.1 Determination of dextran

3.2.1.1 Robert method

Dextran was determined according to Roberts, (1983) and AOAC, (1990). Roberts

copper method determines dextran after polysaccharide precipitation from sugar

solutions by 80 % ethanol. Quantification is made calorimetrically using Phenol-

H2SO4, after a second precipitation with alkaline copper reagent. The amount of

dextran DEm in mg/kg DS in the samples was calculated as follows:

5DE 1011

⋅⋅⋅⋅⋅= FEDC

BAm (3-1)

A Dry substance ( DSw )

B mL of aliquots taken for alcohol precipitation (10 ml)

C mL of solution of alcohol precipitate (25 ml)

D mL of aliquot taken for copper precipitation (10ml)

E mL of final solution of copper-dextran complex (25ml)

F mg/mL dextran (from standard curve)


37

3.2.1.2 Haze method

The determination of dextran in sugar solutions by a modified alcohol Haze method

is conducted according to the ICUMSA method (1994). The test sample is dissolved

in water. Soluble starch is destroyed by incubation with a suitable enzyme (Novo

Termamyl 120L, Novo Industri A/S, Bagsvaerd, Denmark). Protein is removed by

precipitation with trichloroacetic acid (TCA) followed by filtration with acid-washed

kieselguhr. The dextran haze is produced by diluting an aliquot of the treated, filtered

solution to twice the aliquot volume by the addition of ethanol. The turbidity of the

dextran haze is measured by reading the absorbance in a spectrophotometer at a

wavelength of 720 nm. The method is standardized against a commercially available

dextran.

3.2.2 Microbiological experiments

3.2.2.1 Isolation

Sugar beet, cossettes and raw juice were collected during the practical course (2005)

in the Department of Food Engineering, Technical University Berlin. Samples (10 g)

were token and diluted by sterilized distilled water to 100 g. The serial dilutions of

collected samples (10-1 to 10-6) were made and 1 ml portions of the appropriate

dilutions were pour-plated on MRS-Agar media (De Man et al., 1960). The

cultivated microorganisms were incubated anaerobically for 48 hours at 30±1 °C for

enumeration of microorganisms (Leuconostocs sp) (Beukes et al., 2001).

The MRS-broth (Bouillon) consists of (g/liter): Peptone from casein 10, meat extract

8.0, yeast extract 4.0, D (+)-glucose 20.0, di-potassiumhydrogenphosphate 2.0,

Tween 80 1.0, di-ammoniumhydrogencitrate 2.0, sodium acetate 5.0, magnesium

sulfate 0.2 and Manganese sulfate 0.04. To provide MRS-Agar media, agar was

added to the mixture.

3.2.2.2 Identification

3.2.2.2.1 Gas and acid formation

MRS-Bouillon with Durham tubes was used to perform this test with Lactic acid

bacteria according to Back, (1994).


38

3.2.2.2.2 Catalase test

The catalase test involves adding of 5 % hydrogen peroxide (H2O2) to a culture

sample or agar slant. If the bacteria in question produce catalase, they will convert

the hydrogen peroxide and oxygen gas will be evolved. The evolution of gas causes

bubbles (Figure 15) as an indicator of a positive test (Back, 1994).

Figure 15: Colony of bacteria mixed in a 5% solution of H2O2 demonstrating a positive and negative test for producing of catalase.

3.2.2.2.3 Gram characteristics (KOH-Test)

After the mixing of the bacterial colony with ca. 3 % potassium hydroxide on a glass

slide, the KOH solution characteristically became very viscous and mucoid with

gram-negative bacteria. The KOH test was only considered positive if stringing

occurred within the first 30 s of mixing of the bacteria in KOH solution. Gram-

positive bacteria suspended in the KOH solution generally displayed no reaction

(absence of stringing) (Halebian et al., 1981).

3.2.2.2.4 Identification by API 50 CHL test

API 50 CHL Medium, intended for the identification of the genus Lactobacillus and

related organisms, is a ready-to-use medium, which enables the fermentation of 49

carbohydrates on the API 50 CHL strip to be studied.

A suspension is made in the medium with the microorganism to be tested and each

tube of the strip is inoculated. During incubation, carbohydrates are fermented to

acids, which produce a decrease in pH, detected by color change of the indicator. The

results make up the biochemical profile of the strain and are used in its identification

or typing.


39

3.2.2.2.5 L/D-Lactic acid test

For the determination of D- and L- lactic acid, the UV- method, cat. no.

11112821035 from R-Biopharm AG was used. In the presence of D-lactate

dehydrogenase (D-LDH), D-lactic acid (D-lactate) is oxidized to pyruvate by

nicotinamide-adenine dinucleotide (NAD). The oxidation of L-lactic acid requires

the presence of the enzyme L-lactate dehydrogenase (L-LDH). The equilibrium of

these reactions lies on the side of lactate. By trapping pyruvate in a subsequent

reaction catalyzed by the enzyme glutamate-pyruvate transaminase (GPT) in the

presence of L-glutamate, the equilibrium can be displaced in favour of pyruvate and

NADH. The amount of NADH formed in the above reactions is stoichiometric to the

amount of D-lactic acid and of L-lactic acid, respectively. The increase in NADH is

determined by means of its light absorbance at 334, 340 or 365 nm.

3.2.3 Crystallization experiments

3.2.3.1 Measurement of growth rate of sucrose crystals

A laboratory crystallization unit was built according to Wittenberg, (2001). The unit

scheme is given in Figure 16. A double glass wall crystallizer was equipped with

stirrer, automatic measurement devices for dry substance content (Refractometer

IPR2, Schmidt and Haensch) and temperature as well as temperature control. Data

processing was realized on a central computer unit. Dry substance and temperature

were determined every 120 s. Batch isothermal crystallization experiments were

carried at constant temperatures (60, 65 and 70 °C) by seeding syrup with a

supersaturation of 1.15. The experiment was stopped, when the supersaturation had

reached 1.05. Varying admixtures of different dextran fractions (T40, T500 and

T2000) at different concentrations (500 – 5000 mg/kg DS) were used.


40

3

2

4

5

T

M

T +/-

1

Control unit

Central ComputerCrystallizer

Thermostat

Figure 16: Pilot scheme of laboratory crystallization device

1- Thermometer 2 - Automatic Stirrer 3- Double glass wall 4- Cooling water

5- Refractometer (IPR2, Schmidt and Haensch)

3.2.3.1.1 Required amount of dextran and seed

The amount of dextran DEm in mg added to the crystallizing charge was calculated

according to Equation (3-2).

bam ⋅=DE (3-2)

with

100/DSSo wma ⋅=

( )DEW,DE 100/100# wmb −⋅=

Som Total mass of solution in (kg)

DSw Dry substance content in (%)

W.DEm Water content of dextran in %

DE#m Target dextran concentration in mg/kg DS

After reaching the required temperature, the solution in the crystallizer was seeded

with sucrose crystals (size 200 µm) using a syringe. The amount of seed Seedm in g

was calculated as follows:

http://dict.leo.org/se?lp=ende&p=/Mn4k.&search=double

http://dict.leo.org/se?lp=ende&p=/Mn4k.&search=wall


41

3

f

iCryMaSeed ⎟⎟

⎠

⎞⎜⎜⎝

⎛⋅⋅=

ddwmm (3-3)

Mam Mass of the massecuite in g

Cryw Final crystal content in massecuite

id Initial size of the crystal (200 µm)

fd Final size of the crystal

A sample was taken before the seeding point to carry out the first image analysis for

the evaluation of the solubilization process. Further on, samples for image analysis

were taken every 30 minutes in order to follow the growth of crystals through the

crystallization.

3.2.3.1.2 Calculation of the growth rate of sucrose crystals:

In the present work, crystallization was pursued basing on the change of the

refractometric dry substance content and the temperature as:

⎟⎟⎠

⎞⎜⎜⎝

⎛⋅

−−⎟

⎟⎠

⎞⎜⎜⎝

⎛⋅

−=Δ

+

+W

1tML,DS,

1tML,DS,W

tML,DS,

tML,DS,CryS, 100100

mw

wm

ww

m (3-4)

Δ Difference

CryS,m Crystallized sucrose mass

t Time (min)

Ww Mass of water which can be calculated as follow:

( )100

MLDS,SoSoW

wmmm

⋅−= (3-5)

ML Mother liquor

The following equation helps to calculate the sucrose crystal surfaces area:

Cry3/2

CryACry nmfA ⋅⋅= (3-6)

CryA Total crystal surfaces in m2

Af Form factor (0.0423) according to Austmeyer, (1981)


42

Crym Crystal mass

ynCr Number of crystals

The growth rate of sucrose crystals (G) in min)m/( 2 ⋅g was calculated as follows

⎟⎟⎠

⎞⎜⎜⎝

⎛⋅⎟⎟

⎠

⎞⎜⎜⎝

⎛ Δ=

Cry

CryS, 1At

mG (3-7)

3.2.3.2 Dynamic viscosity

A capillary viscometer (Schott-Geräte, capillary no. 532 20/II, Apparatus no.

102221) was used. This viscometer is suitable to determine the kinematic viscosity of

Newtonian liquids according to DIN 51 562, part 1. The instrument constant K is

valid for the survey of the meniscus passage with stands type AVS/S, AVS/S-HT and

AVS/SK from Schott. It was evaluated as K = 0.1028 mm2/s2.

The kinematic viscosity v (mm2/s) can be calculated using the instrument constant in

the equation:

tv ⋅= K (3-8)

t Flow time in seconds corrected according to DIN 51 562, part 1.

The dynamic viscosity η (mPa.s) was calculated by:

ρνη ⋅= (3−9)

where

ρ Solution density

3.2.3.3 Crystal morphology and surface topography

To examine of sucrose surface topography, a scanning electron microscope (SEM,

Hitachi S-2700) was used. After crystallization experiments, the mixture of crystals

and mother liquor (massecuite) was centrifuged and washed by water during

centrifugation, then dried in an oven at 70°C for 3 h.

3.2.3.4 Image analysis

For accurate estimation of the crystallization speed not only the sucrose crystals

quantity but also the beginning area of the sugar crystals is needed. Here, an image

analysis system (IMAGE), which enables a calculation of the middle diameter and

fractionation of sugar crystals was applied (Wagner, 2003).


43

3.2.4 Statistical analysis

Three-way ANOVA were considered in the present work to elucidate the main

effects of the methods of dextran determination with respect to concentrations and

molecular weights in addition to the interaction of these factors. At the same time the

concentration, the molecular weight factors and their interaction were considered in

each method of dextran determination. The homogeneity of variances was assumed

and Tukey-HSD was used at a significant of 0.05. The least sum of squares criterion

was applied to detect the best method for determination of dextran on the light of

molecular weights considered. These analyses were executed using SPSS-Package

release 9 (SPSS-Inc. 1998).

Results and discussion

44

4 Results and discussion

4.1 Sensitivity and accuracy of different methods for the determination of dextrans of varying molecular mass

In this work the sensitivity and accuracy of some common dextran determination

methods is examined in order to select the most appropriate for application in sugar

factories. In this context it must be mentioned that some important considerations are

required to be used in any method used in routine analysis of the sugar factories

according to (Singleton, 2002):

Accurate: The test needs to be as accurate and reproducible as possible within

the constraints of time and simplicity.

Specific: It must quantitatively detect total dextran and must not be biased

towards different molecular weights

Rapid: The test must be able to cope with the high throughput of samples

arriving at factories and allow process decisions to be made quickly.

Simple: To circumvent special training and employment of skilled staff for

repetitive testing the test must be easy to perform. To reduce sample time

sample preparation should also be straightforward and quick.

4.1.1 Robert’s Copper method sensitivity

It is very important to determine the amount of dextran in solution, performing a

standard curve with pure dextran solutions. Figure 17 shows the standard curves of

different dextran molecular fractions: T40, T500 and T2000. The results indicate that

the difference between low and high dextran molecular mass fractions increased

beginning at a dextran concentration ≥ 0.09 mg/ml. However, no significant

differences between all dextran standards were observed except at high dextran

concentrations.


45

R2 = 0.9687

R2 = 0.9933

R2 = 0.9863

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 0.02 0.04 0.06 0.08 0.1 0.12

Dex. conc. (mg/ml)

Abs

.(485

nm)

T40 T500 T2000

Figure 17: Standard Curves of dextran T40, T500 and T2000 by copper complex method.

From the mean of dextran standard curves, the dextran concentrations can be

calculated as follows:

6663.7033.0−

=amDE (4-1)

where

DEm Concentration of dextran (mg/ml)

a = Sample extinction at 485nm.

From the present results, it can be indicated that the standard curve of dextran T2000

matched the measurement data for all molecular mass fractions better than other

curves, whereas R2 for T40 and T500 standard curves were 0.9863 and 0.9687,

respectively, while for T2000 was 0.9933. Consequently, in practical applications the

standard curve for T2000 was applied for all measurements to simplify the

calibration of measurement process, since it is an average of the other molecular

weight curves.


46

Figure 18 to Figure 20 show the patterns of dextran measured by the Robert’s

method and its recovery percentages in relation to the actual dextran added to sucrose

solutions for three different molecular mass fractions of dextran (T40, T500 and

T2000). The results indicated that the measured dextran was highly correlated with

actual ones (R2 > 0.993) whereas the recovery percentage exhibited variable

correlations (-0.891, 0.483 and 0.95 for T40, T500 and T2000, respectively). The

values of recovery percentage were testified against 100 % using least sum of square

referring to T2000 treatment to be the most accurate since its least sum of squares

was the lowest (109.64).

The patterns of variations in the recovery percentages under the effects of dextran

concentrations and dextran molecular weights were considered. The latter factor and

interaction with the concentration were significant (P < 0.01). The concentrations as a

factor have an insignificant effect on the recovery percentages (P = 0.062). Regarding

the measured dextran another pattern of variation was recorded since the

concentration and molecular weight main effects were significant (P<0.0001) with

insignificant interaction (P = 0.053) referring to their independence.

The results indicated that, at especially at low concentrations of the low molecular

dextran fraction T40 Robert’s method overestimated the dextran concentration. At

medium and high concentration recovery was satisfying for all dextran fractions.


47

R2 = 0.9974

0

500

1000

1500

2000

2500

500 1000 1500 2000 2500Dextran present (mg/kg DS)

Dex

.mea

sure

d (m

g/kg

DS

)

0

20

40

60

80

100

120

Dex

t.rec

over

y (%

)

Dex.measured (mg/kg DS) Dex.recovery (%)

Figure 18: Sensitivity of copper complex method to determine dextran T40 in sucrose solutions

R2 = 0.9952

0

500

1000

1500

2000

2500

500 1000 1500 2000 2500

Dextran present (mg/kg DS)

Dex

t. m

easu

red

(mg/

kg D

S)

0

20

40

60

80

100

120

Dex

. rec

over

y (%

)

Dex.measured (mg/kg DS) Dex. recovery (%)



48

R2 = 0.9952

0

500

1000

1500

2000

2500

3000

500 1000 1500 2000 2500


Dex

.mea

sure

d (m

g/kg

DS

)

0

20

40

60

80

100

Dex

.reco

very

(%)

Dex.measured (mg/kg DS) Dex.recovery (%)


4.1.2 Haze method sensitivity

Haze method reflects a pattern of variation in dextran measured similar to that of

Roberts’s method whereas the recovery pattern was different (Figure 21 to Figure

23). The recovery percentage increased with the increasing of dextran concentrations

for the three molecular weights considered (R = 0.96, 0.89, 0.74 for T40, T500 and

T2000 respectively). Testifying recovery percentages against 100 % with the least

sum of square (LSS) criterion, the Haze method was with the best for the molecular

weight of T2000 (Figure 23).

In the Haze method, dextran concentration and molecular weight factors and their

interactions were found to be significant in their effects on the recovery percentages

(P < 0.0001), which points at a dependent relationship between the two factors.

Similar pattern of significance for the estimated dextran T500 and T2000 were

recorded (P<0.0001). The accurate of Haze method with low dextran molecular mass

was shown in Figure 21. The results show the dextran recovery at concentrations

lower than 1000 mg/kg DS and at concentrations above 1500 mg/kg DS of T40 were


49

not accurate compared with the dextran recovery of experiments with T500 and

T2000.

R2 = 0.9875

0

500

1000

1500

2000

2500

3000

3500

4000

500 1000 1500 2000 2500

Dextran present (mg/kg Ds)

Dex

. mea

sure

d (m

g/kg

DS

)

0

20

40

60

80

100

120

140

160

180

Dex

. rec

over

y (%

)

Dex. measured (mg/kg DS) Dex. recovery (%)


R2 = 0.9984

0

500

1000

1500

2000

2500

3000

3500

500 1000 1500 2000 2500Dextran present (mg/kg DS)

Dex

. mea

sure

d (m

g/kg

DS

)

0

20

40

60

80

100

120

Dex

. rec

over

y (%

)




50

R2 = 0.9875

0

500

1000

1500

2000

2500

3000

500 1000 1500 2000 2500


Dex

. mea

sure

d (m

g/kg

DS

)

0

20

40

60

80

100

120

Dex

. rec

over

y (%

)


Figure 23: Sensitivity of haze method to determine dextran T2000 in sucrose solutions

According to the recovery patterns (Figure 18 to Figure 23) and the least sum of

square (LSS) criterion, (Roberts method, 179, 553, 109, Haze method, 8428, 2535,

682 for T40, T500 and T2000 respectively), the Robert’s method can be considered

to be more precise. However, the overall factors including the method, like

concentration, molecular weight and their interaction were considered to discover

another pattern of variation in recovery percentages and determined dextran. For the

recovery percentages, all the factors, including the method and their interactions are

significant (P < 0.0001) whereas for the estimated dextran the method has no main

effect (P = 0.348).

From the above mentioned discussion, the sensitivities of Robert’s and Haze method

for determination of different dextran molecular mass fractions are summarized in

Table 7. The data demonstrate that the Robert’s method was more accurate for

different dextran molecular mass fractions than the Haze method. Furthermore, the

Haze method was too inaccurate to be genuinely useful as a dextran analysis method

especially, with low dextran molecular mass fractions. These results are consistent

with results from (Clarke et al., 1986; Curtin and McCowage, 1986; DeStefano and

Irey, 1986; Sarkar and Day, 1986).


51

Table 7: Conclusion of evaluation of Roberts and Haze methods to determine different dextran molecules

Dextran molecules

T40 T500 T2000

Dextran levels

Haze Copper complex

Haze Copper complex

Haze Copper complex

≤ 500 mg/kg DS - + + +++ + +++

1000-2000 mg/kg DS + ++ ++ + ++ +++

≥ 2000 mg/kg DS - +++ ++ + ++ +++

- not sensitive, + low sensitive, ++ sensitive, +++ high sensitive

4.2 Microbial sources of dextran an identification of relevant microorganisms in sugar factories

In this chapter, the isolation and identification of Leuconostoc mesenteroides from

sugar beet factories is described. In contrast to other lactic acid bacteria, Leuconostoc

sp. can tolerate fairly high concentrations of salt and sugar (up to 50% sugar) and

initiates growth in vegetables more rapidly over a range of temperatures than any

other lactic acid bacteria. It produces carbon dioxide and acids which rapidly lower

the pH (Battcock and Ali, 1998). Furthermore, it produces high molecular mass

polysaccharides such as dextran, which causes many problems during the industrial

process in sugar factories.

Climatic conditions and harvesting method as well as storage place in sugar factories

are important factors influencing the deterioration of sugar beet by microorganisms.

The raw materials are vulnerable to enzymatic deterioration or microbial action

(Figure 24). Usually, in this regard hand-harvested sugar beet has a lower

deterioration rate after harvest than machine-harvested, whereas, the machine causes

a cutting of sugar beets through the harvesting. This means, they are vulnerable to

attack rapidly from the microorganisms.


52

Figure 24: Deteriorated sugar beet after harvesting

The enumeration of microorganisms in the collected samples (sugar beet, cossettes

and raw juice from beet sugar factory) after incubation on MRS-Agar at 30°C for 48

hours was performed. The mean counts of colonies were 2.5*103 cfu/g of the good

sugar beet sample. On the other hand, they were 34*103 cfu/g in the deteriorated

sugar beet.

Also, the results indicate that the count of microorganisms was reduced in cossettes

samples. Due to the more efficient washing procedure and lower initial soil loading in

beet processing there is less extraneous material brought into the factory. The mean of

total microorganism colonies in these beet sugar cossettes were almost 6*102 cfu/g

and they were found only in the dilutions 10-1 and 10-2.

It was observed that the number of microorganisms increased in raw juice after

extraction process. In the first day of the sugar beet manufacture, the number of

colonies in raw juice was 23*104 cfu/g. On the other hand, in the second and third

days, the microbial count increased to 4*106 and 3.5*105 cfu/g, respectively. This

increase was may be due to starting microbial contamination and an underdosed

usage of antimicrobial agents.

Figure 25 shows the relationship between growth of microorganisms and storage time

of sugar beet raw juice at room temperature. The counting of microorganisms in the

first time was 3.35*103 cfu/g, but after 1, 2, 3 and 4 hours it was 3.35*103, 6.9*103,

2.2*104 and 6.5*104 cfu/g, respectively. The suitable sugar concentration, pH and

temperature of raw juice apparent in an industrial surrounding contribute greatly to

the increase in the rate of growth of microorganisms. Day, (1992) reported similar


53

results. Also he described that during raw juice extraction in the diffuser a decrease of

the cossettes temperatures can occur, if a large percentage of damaged beets are

processed or the usual capacity of the extraction device is exceeded. Consequently,

the bacterial loading increases.

0

10000

20000

30000

40000

50000

60000

70000

0 1 2 3 4

Time (hours)

Num

ber o

f mic

roor

gani

sms

col

onie

s in

(cfu

/g)

Figure 25: Microorganisms growth in sugar beet raw juice at room temperature

The colonies morphology of isolated strains and cultural conditions are shown in

Appendix 1. The results show that 14 microorganism strains were isolated from the

collected samples Classical identification revealed the presence of seven bacteria

strains. The other colonies were yeasts and fungi. In this study, the identification of

Leuconostoc mesenteroides subsp. mesenteroides, which produces the dextrans, was

in the focus of interest.

On the basis of classical phenotypic analysis, all of isolated bacteria strains were

Gram-positive and gas forming (except S3 no gas formation). However, S1, S5, S12,

S13 and S14 were catalase negative while S2, S4 were catalase positive. In addition,

all strains produced acid from glucose (See Appendix 2).

The fermentation pattern of carbohydrates gives other keys for strain identification.

Fermentation patterns or profiles and biochemical characteristics were determined


54

using API 50 CHL (BioMérieux) for Leuconostoc ssp. The fermentation pattern

allows the isolate to be identified comparing the pattern to the characteristics of the

type strains. Even in optimized conditions, the change of color may sometimes take

so long for a given strain that it is difficult to conclude whether there is a positive or

negative reaction. Therefore, in this work, follow-up the color changing has taken

several days (five days). Identification results were obtained with the aid of the API

identification table (Bio Merieux) (see Appendix 3). From the five isolated bacteria

strain, the strain S1 belonged to the species Leuconostoc mesenteroides subspecies

mesenteroides/dextranicum. Another isolated strain could not be identified.

Another key to confirm the isolated Leuconostoc m. is the determination of produced

D-/L-lactate. The optical isomer of lactic acid was determined by using a D-/L-

lactate dehydrogenase kit. Our results indicated that the amount of D-Lactic acid

which produced by strain 1 (S1) was 7.367 g/L. whereas the amount of L-Lactic acid

was 0.387 g/L. In this case, the L & D – Lactic acid test confirms that the tested

strain was Leuconostoc m.

The morphology of strain 1 (S1) was examined by using scanning electron

microscopy (S-4200; Hitachi). Figure 26 shows that the strain 1 cells are spherical

and resemble very short bacilli with rounded ends. The size of cells is approximately

0.6 to 1.0 µm. Cells are arranged in single or pairs.

These results are in agreement with Lonvaud-Funel, (1999). He reported that the

Leuconostoc cells are often in short chains in nutrient media during the active growth

phase, whereas in their natural environment and more stressful conditions the chains

zone. From the abovementioned our results can be confirm the strain 1 (S1) is

Leuconostoc mesenteroides subsp mesenteroides which produced dextran molecules

in sugar industries.


55

Figure 26: Scanning electron micrograph of the isolated Leuconostoc mesenteroides.

Lonvaud-Funel, (1999) reported that the Leuconoistc cells are often in short chains in

nutrient media during the active growth phase, whereas in their natural environment

and more stressful conditions the chains zone.

From the abovementioned our results can be confirm that the strain 1 (S1) is

Leuconostoc mesenteroides subsp mesenteroides which produces dextran molecules

in sugar industries.

4.3 Levels of dextran contents in different sugar beet factories

The dextran level in beet juices and syrups depends on many factors such as

agricultural and climatic conditions, harvesting methods, length of time between

field and factory, length and condition of beet storage as well as the efficiency of all

manufacturing operations. The juices and syrups under study came from Egyptian

factories located in northern (Factory1 and 2) and middle Egypt (Factory 3).

Figure 27 exemplifies the progression of dextran concentrations along the production

process. The dextran levels in factories 1 (F1) and 2 (F2) were significantly higher

than in factory 3 (F3), which was mainly due to the fact that F3 had a shorter beet

campaign of only one month. In addition, the cultivation of sugarbeet in middle

Egypt is new compared to the cultivation in northern Egypt. The levels of dextran in

cossettes were 878, 1343 and 747 mg/kg DS for factories 1, 2 and 3, respectively.

During the extraction the dextran content increased by 4 %, 19 % and 17 % for F1,

F2 and F3, respectively. This means that the microbiological control in F1 was better


56

than in F2 and F3. In contrast, the lime-carbon dioxide juice purification reduced the

dextran content in juice to 40%, 50% and 60% for F1, F2 and F3 respectively by

removing insoluble dextrans during filtration steps. As mentioned above, soluble

dextran may lead to problems in the evaporation and crystallization, and

consequently to sugar loss.

The dextran levels in the final production steps (white sugar and molasses) of the 3

Egyptian factories are shown in Figure 28. The levels of dextran in white sugar

(sugar a) were 38.50, 62.38 and 21.00 mg/kg DS for factories 1, 2 and 3,

respectively. On the other hand were 314, 472 and 405 mg/kg DS in molasses for

factories 1, 2 and 3, respectively.

0

300

600

900

1200

1500

Cossettes Raw juice Clarified juice Thick juice

Samples

Dex

tran

cont

ent (

mg/

kg D

S)

F1 F2 F3

Figure 27: Dextran levels at 3 Egyptian factories (F1, F2 and F3) during the production processes


57

0

100

200

300

400

500

Sugar a Molasses

Final products

Dex

tran

cont

ent (

mg/

kg D

S)

F1 F2 F3

Figure 28: Dextran levels in final products (white sugar and molasses) of 3 Egyptian factories (F1, F2 and F3)

The level of dextran in molasses was higher than in the resulting product (sugar).

This shows that the major amount of dextran was lost to the molasses during the

washing process to the molasses.

Based on the cluster analysis of some characters, the relationship between the 3

Egyptian factories was elucidated in Figure 29. The result indicated a correlation

between factory 1 and factory 3. This correlation may be due to the use of a good

sugar beet during the campaign, which leads to reduce the dextran levels during the

production processes. On the other hand, the use of the deteriorated sugar beets in

factory 2 caused an increase of dextran levels during the processes.


58

Linkage Distance

FACTORY2

FACTORY3

FACTORY1

2,2e5 2,4e5 2,6e5 2,8e5 3e5 3,2e5 3,4e5 3,6e5 3,8e5 4e5

Figure 29: The relationship between the three factories based on cluster analysis of some characters (Single linkage squared Euclidean distance)

4.4 Quality of factory final products and their relationship to the levels of dextran during different industrial periods

An increase of the molasses purity and sugar loss during the development of

industrial periods was observed in sugar beet factories in Egypt and many another

countries, especially at the end of the industrial season. In this chapter, we will

highlight on the correlation between the increase of sugar loss and the presence of

dextran during different industrial periods as well as during the production process

itself.

Figure 30 shows the relationship between molasses purity and the industrial periods.

The results indicate that molasses purity increased steadily during the campaign and

at the end of the industrial season, the dextran concentration was enhanced by 4.6 %.

Whereas, the dextran level in syrup increases from about 500 mg/kg DS during

March to 1200 mg/kg DS at the end of May. This value is in good accordance with

literature data given in Table 5 as starting levels of dextran were around 1000 mg/kg

DS. In the European factories, the presence of raffinose may lead to similar results,

whereas it increases at the end of the season.


59

54

55

56

57

58

59

60

Marc

h 1 2 3 4

Apri

l 1 2 3 4

May

1 2 3 4

Industrial period

Mol

asse

s pu

rity

%

Molasses purity (Mean) Molasses purity

+1.35%

+4.63%

0%

Figure 30: The relationship between molasses purity and the industrial periods. (Mean = mean of molasses purity in month)

The results obtained from the sugar factory 2 showed that despite an increase in the

percentage of sucrose in beet cossettes from 17.70 % in March to 18.34 % in May,

the sugar extraction quality decreased from 85.74 % to 81.21 % and the sugar loss in

molasses increased from 2.5 to 2.8 (kg/100kg beet) in March and May, respectively.

This is mainly based on the deterioration of the sugar beet crop by microorganisms.

Especially the presence of Leuconostoc mesenteroides leads to an increase of the

dextran levels during the production processes.

Figure 31 shows the dextran levels in raw juice, clarified juice and thick juice during

different industrial periods of factory 2. The results illustrated that the dextran levels

were higher in May than in March in all samples. The dextran amount increased from

March to May by 700 mg/kg DS to a range of 1000-1400 mg/kg DS, by 400 mg/kg

DS to a range 650-870 mg/kg DS and by 500 mg/kg DS to range of 740-1150 mg/kg

DS of raw juice, clarified juice and thick juice, respectively.


60

0

200

400

600

800

1000

1200

1400

1600

Raw juice Clarified juice Thick juice

Samples

Dex

tran

cont

ents

(mg/

kg D

S)

March 12.May 21.May 28.May

Figure 31: Influence of the industrial period on the dextran amount during production processes

Also, the amount of dextran in different sugar production steps (sugar a, b, d, g) and

in molasses was increased at the end of the industrial season (Figure 32). The results

indicated that the dextran levels increased with the developing crystallization steps.

They were lower in the initial crystallization stage and then increased gradually to

reach 103 mg/kg DS in the last crystallization stage (g-sugar, after the last production

step). These results confirm the importance of dextran effects during the later stages

of crystallization processes. In addition, the increase of sugar beet deterioration in the

end season causes an increase in the dextran levels in all crystallization stages.


61

0

50

100

150

200

250

300

350

400

450

500

March 12.May 21.May 28.May

Industrial period

Dex

tran

cont

ent (

mg/

kg D

S)

Sugar a Sugar b Sugar g Sugar d Molasses

Figure 32: Dextran levels in sugar and molasses production of different industrial periods

The result in Figure 33 illustrates the relationship between dextran levels in syrup

and dextran levels in sugar during different crystallization steps. The dextran level in

white sugar and after product were 32 and 103 mg/kg DS, respectively, when the

dextran in syrup was 500 mg/kg DS, respectively. Furthermore, an increase in the

dextran level in syrup to 1100 mg/kg DS, increase the dextran amount of white sugar

and after product to 85 and 170 mg/kg DS, respectively, was observed. Lower

crystallization temperatures and the adsorptive properties of dextran contribute

greatly to the increase of dextran levels in the after product crystallization step.

Figure 34 shows the partitions of dextran in sugar a, b and molasses during March

and May. The dextran levels in all tested ‘a-sugar’ samples were lower than dextran

contents in ‘b-sugar’. The results indicated that about 45-50 % of dextrans in magma

tend to the molasses. However, about 17-20 % remain in sugar g (after product),

which normally is recycled to the first stage of crystallization. In addition, the

dextran levels increase with the increase of crystallization stages.


62

From the present results, it can be confirmed that the dextran concentrates on the

crystal surfaces due to the high ability of dextran to adsorb on surfaces. Furthermore,

the amount of dextran in sugar product depends on the amount of dextran in syrup

and the quality of the affination process.

Sugar a

Sugar b

Sugar g

Sugar d

0

20

40

60

80

100

120

140

160

180

500 600 700 800 900 1000 1100 1200

Dextran in syrup (mg/kg DS)

Dex

tran

cont

ent (

mg/

kg D

S)

Figure 33: The relationship between the dextran content in syrup and different produced sugars

March

6% 11%

19%

49%

15%

Sugar a Sugar bSugar d Sugar g (After product)Molasses

May

10%

13%45%

17%

15%

Sugar a Sugar bSugar d Sugar g (After product)Molasses

Figure 34: The partitions of dextran during different crystallization stages at different industrial periods


63

4.5 Influence of dextran concentrations and molecular fractions on the rate of sucrose crystallization in pure sucrose solutions

There are different methods to perform the crystallization experiment and follow the

growth rate of sucrose crystals during the crystallization process. In laboratory

isothermal crystallization experiments (as described under 3.2.3.1) the influence of

dextran on sucrose growth was studied without interference of other factors.

The effect of different dextran concentrations and molecular weight fractions on

crystallization rate in sucrose solutions will be discussed ahead to demonstrate the

complexity of the problem and to elucidate the methodology of performed

crystallization experiments. Based on first experiments the influence of temperature

on crystallization rate in presence and absence of dextran will be balanced, too. All

presented results finally demonstrate the sensitivity of the crystallization process on

dextran concentration in the system. From this, the consequences of dextran presence

for industrial crystallization units can be derived.

Figure 35 shows the dry substance contents of the mother liquor ( MLDSw , ) during the

crystallization (ϑ = 60°C) for a pure sucrose solution and after addition of 500, 1500

and 2000 mg/kg DS dextran T40. The results indicate that the dry substance of

mother liquor decreases during the crystallization process. The figure shows that

concentrations of up to 500 mg/kg DS of dextran T40 in the sucrose syrup have only

little effect on the dry substance decrease. However, concentrations between 1500

and 2000 mg/kg DS of dextran with the same molecular weight however give effect

on the exhaustion of mother liquor.

The growth rates of sucrose crystals calculated from the data shown in Figure 35

during the crystallization process for pure sucrose solution after addition of dextran

T40 of different concentrations (500, 1500 and 2000 mg/kg DS) at 60°C are shown

in Figure 36. Only small differences in the crystallization rates of control (without

dextran) and with addition of 500 mg/kg DS of dextran T40 were observed. On

contrary, the admixture of higher concentrations of T40 (1500-2000 mg/kg DS)

causes a serious retardation of crystallization rate and thus represents a danger for

crystallization process economy, in particular if industry-relevant supersaturations

above 1.1 are regarded. For example, at SS 1.12 the crystallization rate with 2000

mg/kg DS T40 is almost 15 % lower compared to the control experiment. On the


64

other hand, at supersaturation 1.13 (as indicated in the diagram), the growth rate is

reduced by 7.4 %, 9 % and 18 % after dextran T40 at concentrations of 500, 1500

and 2000 mg/kg DS, respectively, was added.

75

75.2

75.4

75.6

75.8

76

76.2

76.4

76.6

76.8

0 20 40 60 80 100 120 140 160

Time (min)

DS

in (%

)

Control 500 mg/kg DS 1500 mg/kg DS 2000 mg/kg DS

Figure 35: Dry substance content of mother liquor in isothermal crystallization experiments at 60 °C in pure sucrose solution and after addition of 500, 1500 and 2000 mg/kg DS of dextran

T40

0

0.5

1

1.5

2

2.5

3

3.5

4

1.05 1.06 1.07 1.08 1.09 1.1 1.11 1.12 1.13 1.14

Supersaturation (SS)

G (g

/ m

2 .min

)


Figure 36: Growth rate of sucrose crystals in isothermal crystallization experiments at 60 °C in pure sucrose solution and after addition of 500, 1500 and 2000 mg/kg DS of dextran T40


65

Figure 37 illustrates the relationship between the dry substance content and the time

of the crystallization process for pure sucrose solutions (control) and after addition of

dextran T500 (500, 1500 and 2000 mg/ kg DS) at 60 °C. An increase of molecular

weight of dextran from 40.000 (T40) (s. experiments above) to 500,000 (T500) at

60°C caused an increase of crystallization time. A relevant effect of dextran T500

was observed with concentrations higher than 500 mg/kg DS.

Figure 38 shows the effect of dextran T500 at concentrations of 1500 and 2000

mg/kg DS on the growth rate of sucrose crystals at 60°C. This figure indicates that

the crystallization rate regarding a supersaturation of 1.13 in the sugar solution

without dextran (control) is the same as the crystallization rate at supersaturations of

1.134, 1.135 and 1.138 in presence of 500, 1500 and 2000 mg/kg DS of dextran

T500, respectively. Also, a stronger effect of dextran T500 than dextran T40 can be

observed at 60°C. As it can be seen from Figure 38 that the crystallization rates at

60°C at a supersaturation of 1.13 are reduced by 7 %, 10 % and 20 % in presence of

500, 1500 and 2000 mg/kg DS dextran T500, respectively.

75

75.2

75.4

75.6

75.8

76

76.2

76.4

76.6

76.8

0 20 40 60 80 100 120 140 160 180

Time (min)

DS

in (%

)


Figure 37: Dry substance content of mother liquor in isothermal crystallization experiments at 60 °C in pure sucrose solution and after addition of 500, 1500 and 2000 mg/kg DS of dextran

T500


66

0

0.5

1

1.5

2

2.5

3

3.5

4

1.05 1.06 1.07 1.08 1.09 1.1 1.11 1.12 1.13 1.14


G (g

/m2 .m

in)


Figure 38: Growth rate of sucrose crystals in isothermal crystallization experiments at 60 °C in pure sucrose solution and after addition of 500, 1500 and 2000 mg/kg DS of dextran T500

The results from Figure 37 and Figure 38 clearly demonstrate that the reduction of

growth rate at technologically relevant supersaturations is even more than described

in Figure 35 and Figure 36 (dextran T40). That means, if in a sugar house a thick

juice contains 1500 mg/kg DS dextran T500, the crystallization period will be

extended by up to 20 %.

Further experiments on the influence of high dextran concentrations were carried out

with dextran T2000. The mother liquor dry substance contents in the presence of

500, 1500 and 5000 mg/kg DS of dextran T2000 are shown in Figure 39. The time

till a supersaturation of 1.05 was reached increased by 21 %, 36 and 41 % after 500,

1500 and 5000 mg/kg DS dextran T2000 addition, respectively.

The growth rate of sucrose crystals at 60 °C and a supersaturation of 1.13 were 2.7,

2.4, 1.8 and 1.3 min)g/(m2 ⋅ for control and after dextran T2000 addition of 1500

and 5000 mg/kg DS, respectively (see Figure 40). Here the crystallization rates were

reduced by 12, 33 % and 51 % after addition of 500, 1500 and 5000 mg/kg DS

dextran T2000


67

75

75.2

75.4

75.6

75.8

76

76.2

76.4

76.6

76.8

0 20 40 60 80 100 120 140 160 180 200

Time (min.)

DS

in (%

)

Control (60°C) 500 mg/kg DS 1500 mg/kg DS (60°C) 5000 mg/kg Ds (60°C)

Figure 39: Dry substance contents of mother liquor in isothermal crystallization experiments at 60°C in pure sucrose solutions and after addition of 500, 1500 and 5000 mg/kg DS of dextran

T2000

0

0.5

1

1.5

2

2.5

3

3.5

4

1.05 1.06 1.07 1.08 1.09 1.1 1.11 1.12 1.13 1.14

Supersaturation (SS)Control (60°C) 1500 mg/kg DS (60°C)5000 mg/kg DS (60°C) 500 mg/kg DS

G (g

/ m

2 .min

)

Figure 40: Growth rates of sucrose crystals in isothermal crystallization experiments at 60°C in pure sucrose solutions and after addition of 500 - 5000 mg/kg DS of dextran T2000


68

The comparison of the effect of the different molecular mass fractions of dextran on

the degradation of dry substance contents of the mother liquor ( MLDSw , ) during the

crystallization at ϑ = 60°C for a pure sucrose solution and after addition of 1500

mg/kg DS dextran T40, T500 and T2000 is shown in Figure 41. The time till a

supersaturation of 1.05 was reached increased by 36 % compared with a pure sucrose

solution after addition of dextran T2000. In contrast, only a small effect was found

after addition of lower molecular mass fractions of dextran (T40 and T500) under the

same conditions.

Figure 42 shows the crystal growth rates of sucrose were calculated from the data

shown in Figure 41. The growth rate of sucrose crystals was reduced after dextran

T40, T500 and T2000 addition by 9 %, 10% and 33 %, respectively for an

industrially relevant supersaturation of 1.13.

75

75.2

75.4

75.6

75.8

76

76.2

76.4

76.6

76.8

0 20 40 60 80 100 120 140 160 180

Time (min)

DS

in (%

)

Control T40 T500 T2000

Figure 41: Dry substance content of mother liquor in isothermal crystallization experiments at 60 °C in pure sucrose solution and after addition of 1500 mg/kg DS of dextran T40, T500 and

T2000


69

0

0.5

1

1.5

2

2.5

3

3.5

4

1.05 1.06 1.07 1.08 1.09 1.1 1.11 1.12 1.13 1.14


G (g

/ m

2 . min

)


Figure 42: Growth rate of sucrose crystals in isothermal crystallization experiments at 60 °C in pure sucrose solution and after addition of 1500 mg/kg DS of dextran T40, T500 and T2000

4.5.1 Influence of different temperatures on growth rate of sucrose crystals in the presence of dextran

In this chapter, the role of the temperature during the crystallization is studied. As the

dextran molecular fraction T2000 showed the most relevant effect on crystallization

rate of sucrose crystals only results of experiments with this molecular fraction are

presented. Figure 43 shows mother liquor dry substance content at 60, 65 and 70 °C

in pure sucrose solution and after addition of 1500 and 5000 mg/kg DS of dextran

T2000. The results indicate that great differences in the rate of dry substance

reduction were apparent at the regarded temperatures. At a crystallization

temperature of 60 °C the crystallization time was extended by 33 % after the

admixture of dextran (5000 mg/kg DS) compared to the control experiment. At 70 °C

the extension of crystallization time for the same admixture was reduced to only 8 %.

It was observed that the degree of reduction at 70°C was higher than at 65 °C or

60°C considering all treatments. In contrast, the effect of dextran appeared to be

reduced at 70°C and increased at lower temperatures (65 °C and 60°C).

Also, a higher dextran concentration (5000 mg/kg DS) had a stronger effect on the

exhaustion of mother liquor than a concentration of 1500 ppm under the


70

experimental conditions. Hence, longer time of crystallization was required at low

temperatures and high dextran contents.

74.5

75

75.5

76

76.5

77

77.5

78

78.5

79

0 50 100 150 200

Time (min)

DS

in (%

)

Control (60°C) Control (65°C)Control (70°C) 1500 mg/kg DS (60°C) 5000 mg/kg Ds (60°C) T2000-1500ppm (65°C)1500 mg/kg DS (70°C) 5000 mg/kg DS (70°C)

60°C

70°C

65°C

Figure 43: Dry substance content at 70, 65 and 60°C in control and in presence of dextranT2000 (1500 and 5000ppm) during crystallization process

The influence of addition of 1500 and 5000 mg/kg DS of dextran on crystal growth

rates at 60 and 70 °C is shown in Figure 44 and Figure 45. The addition of 1500

mg/kg DS of dextran T2000 reduced the growth rate of sucrose crystals by 33 % at

60 °C, 24 % at 65°C and by 16 % at 70 °C. These results show the same tendency as

results from Ekelhof, 1997; Meade and Chen, (1977). However, growth rates

obtained in own experiments were higher than the former results of Ekelhof. This is

due the use of a fast mixing system during own experiments in contrast to the float

bed (without stirrer) crystallizator applied by Ekelhof. At a supersaturation of 1.13

addition of 5000 mg/kg DS of dextran T2000 reduced the growth rate of sucrose

crystals by 51 % and 24 % at 60 °C and 70 °C, respectively (see Figure 44 and

Figure 45).

The results indicate that higher temperatures (e.g. 70 °C) reduce the effect of dextran

on the crystallization rate especially in case of high levels of dextran in the


71

massecuite. Furthermore, crystallization temperature at 70°C improves the mother

liquor diffusion and consequently, the growth rate of sucrose crystals.

0

1

2

3

4

5

6

7

1.05 1.06 1.07 1.08 1.09 1.1 1.11 1.12 1.13 1.14Supersaturation (SS)

G (g

/m2.

min

)

Control (60°C) 1500 mg/kg DS (70°C) Control (70°C)

1500 mg/kg DS (60°C) 1500 mg/kg Ds (65°C) Control (65°C)

G (g

/ m

2 .min

)

70°C

60°C

65°C

Figure 44: Growth rate of sucrose crystals at 60, 65 and 70 °C after addition of 1500 mg/kg DS of dextran T2000

0

1

2

3

4

5

6

7

1.05 1.06 1.07 1.08 1.09 1.1 1.11 1.12 1.13 1.14Supersaturation (SS)

G (g

/ m

2 . min

)

Control (70°C) Control (60°C)5000 mg/kg DS (70°C) 5000 mg/kg DS (60°C)

Figure 45: Growth rate of sucrose crystals at 60 and 70 °C after addition of 5000 mg/kg DS of dextran T2000


72

4.6 Elucidation of crystallization kinetics in presence of dextran molecules

From the last chapter it could be seen that dextran has a strong influence on the

growth rate of sucrose crystals, especially considering high molecular fractions of

dextran. In this chapter, reasons for the influence of dextran on sugar crystallization

are discussed and illustrated with additional measurement results.

Figure 46 shows a model of sucrose crystallization adapted from Ekelhof, (1997) and

Thelwall, (2002). Around the crystal surface a boundary layer exists which consists

of three parts. The first boundary layer at the crystal surface is the reaction layer,

which has the least supersaturation (CG) as sugar had been removed from the solution

and deposited on the crystal. The second boundary layer is the main diffusion layer.

Here, there are two different flows in opposite directions. The first one is non-sugar

material and water from the exhausted reaction layer moving away from the surface

and back to the main mother liquor. The other flow is fresh syrup, which is moving

towards the crystal surface. The third and final layer, bulk of solution, is the main

body of the mother liquor and is the source of the sucrose molecules for further

deposition.

The movement of sucrose molecules through the diffusion layer (d2) depends on the

viscosity, which does have an important influence on the crystallization rate,

especially in the second boundary layer, as the higher viscosity will reduce the speed

of the molecular movement to and from the crystal surfaces (compare results in

chapter 4.7).


73

d1d2

d3

δλ

Reaction layer

Surf

ace

area

Diffusion layer

kD kR

Cry

stal

Sucr

ose

solu

tion

cL

cG

cSat NS & Water

Figure 46: The scheme of the two processes of diffusion and surface reaction. 2λ= gap between crystals, λ = half distance between crystals, δ = thickness of the boundary layer, kD = diffusion

coefficient, kR = surface reaction, NS = non-sugar, CL, CG, CSat are the sucrose concentrations of liquor, boundary layer and saturation at the crystal-solution interface.

Also, in impure sucrose solutions (sugar cane or beet syrup), dextran molecules are

adsorbed on the impure particles, consequently, the solution viscosity increases and,

additionally, the movement of sucrose molecules destined to crystal surfaces is

impeded (s. Figure 47). Hence the high molecular weight and the adsorption of

dextran affect the growth rate and the quality of sucrose crystals negatively.

Figure 47: Dextran adsorption on impure particles


74

These model suggestions are in accordance to Cossairt, (1982). He demonstrated that

some impurities are physically adsorbed on the different faces of the crystals in

varying and selective amounts. A portion of lattice sites becomes blocked; thereby it

comes to reduction or retardation of crystal growth.

4.7 Influence of dextran molecule fractions on sucrose solution viscosity

In this work, the solution viscosity is a determining parameter for diffusion

phenomena during crystallization of sucrose. Figure 48 demonstrates the influence of

dextran molecular weight fractions T40, T500 and T2000 on the dynamic viscosity

(η) of sucrose solutions (WDS = 60 %) at different temperatures (50-70°C). At 50°C

the dynamic viscosity of sucrose syrup was 14.35 mPa.s for control. On the other

hand, it was 14.63, 15.26 and 15.39 mPa.s with 1500 mg/kg DS dextran T40, T500

and T2000 additions, respectively. In addition, an increase of solution temperature to

70°C reduced the solution viscosity by 50 %.

6

7

8

9

10

11

12

13

14

15

16

50 55 60 65 70

Temperature (°C)

η m

Pa.

s


Figure 48: Influence of addition of 1500 mg/kg DS of T40, T500 and T2000 dextran on the dynamic viscosity of a 60% sucrose solution between 50 and 70 °C

The results elucidate that at low temperatures dextran fractions T2000 and T500 have

more effect on solution viscosity than T40. However, the differences are reduced at


75

high temperatures. That illustrates the decrease of the diffusion rate in the mother

liquor during crystallization process in presence of high molecular weight fractions

of dextran (Figure 46).

Figure 49 shows the relationship between dextran addition and solution viscosity at

different temperatures. At 50 °C the solution viscosities increased after addition of

500, 1500, 2500 and 5000 mg/kg DS of dextran T2000 by 2.9 %, 6.45 %, 11.70 %

and 24.73 %, respectively. An increase of temperature from 50 to 70°C reduced the

solution viscosity to almost 50 % for all additions.

0

2

4

6

8

10

12

14

16

18

20

Control 500 1500 2500 5000

Dextran contents (mg/kg DS)

η (m

Pa.

s)

50°C 60°C 70°C

Figure 49: Influence of different dextran concentrations (500, 1500, 2500, 5000 ppm) on the dynamic viscosity at different temperatures (50-70°C)

The effects of dextran are not only of importance during the crystallization process,

but also after the centrifugation process and during the drying operation of the

sucrose crystals. These steps as well determine the quality of the resulting sucrose

crystals. Figure 50 shows the crystal surface during the drying process.


76

A f te r A f te r c e n t r i fu g a t io nc e n t r i fu g a t io n

A f te r d ry in g A f te r d ry in g (u n s ta b le s ta te )(u n s ta b le s ta te )

C ry s ta l lin e C ry s ta l lin e p h a s ep h a s e

S a tu ra te d S a tu ra te d s o lu t io ns o lu t io n

S a tu ra te d S a tu ra te d s o lu t io ns o lu t io n

G la s s y G la s s y s ta tes ta te

Figure 50: The crystal surface of a sucrose crystal during the drying process (according to Bunert und Bruhns 1995)

It can be shown that the presence of dextran in the saturated solution on the crystal

surface leads to an increase of solution viscosity. So, heat transfer and consequently

water transport to the air is slow. Hence, an increase of the drying time contributes to

an increase of micro-crystals (fine crystal) due to nucleation on the crystal surfaces.

4.8 Influence of dextran on the morphology and surface topography of sucrose crystals in presence of dextran

4.8.1 Crystal morphology

After the characterization of the influence of dextran on the crystal growth rate, in

this chapter the influence on crystal shape and topography shall be illustrated in this

chapter. In presence of dextran sucrose crystals are often elongated along the c axis

(Figure 51-b). Dextran influences the growth rates of individual crystal faces

(McGinnis, 1982) as described in chapter 2.6.4. In addition, the adsorption properties

of dextran causes the deposition of impure materials on crystal surfaces (Figure 51-

c), which reduces the sugar quality and increases the sugar loss to molasses.


77

cba

+C

+B

+A+C

+B

Figure 51: Effect of dextran on morphology of sucrose crystals, a) Normal crystal, b) Crystal grown in presence of dextran (needle crystal) and c) Impure particles on crystal surface (crystal

grown in technical sucrose solution)

The relationship between the presence of dextran and elongation of crystals (c/b

ratio) was elucidated by Bubnik and Kadlec, (1996). They noticed that the variation

of the c/b ratio is a function of increasing concentration of dextran (s. chapter 2.6.4.).

Results of own experiments showed that the low molecular weight dextran caused an

elongation of crystals, despite the use of stirring during the crystallization process

(see Figure 52). This result is in good agreement with Singleton, (2002). He reported

that dextrans are responsible for crystal elongation along the c axis and that the

smaller molecular weight dextrans are especially involved in this effect. However in

own experiments a stirred crystallizer device and seeding crystals were applied.

Therefore, crystals are only slightly elongated and no pure needle crystal formation

took place.

Without dextran T40-1000ppm T40-2000ppm

Figure 52: An example of influences of dextran T40 on sucrose crystal shape during the crystallization process

4.8.2 Surface topography

Microscopic examinations of the different faces of sucrose crystals were performed

by scanning electron microscopy (SEM). Figure 53 shows the surface topography of


78

a sucrose crystal which was grown in pure sucrose solution at 60°C. It was observed

that the crystal shapes has taken a normal form of a sucrose crystal. On the other

hand, the microscopic examination of a crystal surface shows the fine sugar crystal

on the crystal surface, which developed after the centrifugation process and during

the drying operation (comp. chapter 4.7).

20x 150x

6000x 1000x

Figure 53: Surface topography of a laboratory sucrose crystal grown in absence of dextran at 60°C

Figure 54 shows the effect of the low molecular dextran fraction (T40) on sucrose

crystal shape and the surface adsorption. The crystals shown are formed from 200

µm seed crystals and therefore no needle crystals are formed. In contrast, crystals

nucleated in the crystallizer in the presence of dextran (top left Figure 54) were small

needle crystals, which get lost to molasses easily during centrifugation and washing,

consequently, increasing the molasses purity. Also, an increase of micro-crystals on

the crystal surfaces took place compared with control was observed. So, rough

surfaces and gaps developed on the crystals. In this case, the main affect of low

molecular dextran is changing of crystal form to needle crystal. Consequently, there

is an increase in sugar loss to molasses.


79

6000x 1000x

20x 150x

Figure 54: Surface topography of sucrose crystal grown in the presence of 1500 mg dextran T40 per kg DS at 60 °C

Figure 55 shows the influence of the high molecular fraction dextran molecules

T2000 on sucrose crystal morphology and adsorption surface. It was observed that in

presence of this high molecular fraction of dextran the number of conglomerated

crystals (multiple crystals where two or more crystals have grown together)

increased. A scanning electron micrograph of a sugar crystal shows micro-particles

(sucrose crystals) strongly adsorbed on the crystal surface, which cause a rough

crystal surface with a number of gaps (comp. chapter 4.7). Consequently, non-sugar

and colorant particles can be adsorbed with dextran on the surface during the

technical crystallization process in the sugar factories. Also in this case, the removal

of impurities (non-sugars) is difficult during the washing of the crystals.


80

20x

6000x 1000x

150x

Figure 55: Surface topography of a sucrose crystal grown in the presence of 5000 mg dextran T2000 per kg DS at 60°C

These results confirmed those reported by Cuddihy et al., (2000); Jimenez, (2005);

Karagodin, (1998); Rauh et al., (1999). They found that the increase of sugar

crystallization time causes the massecuites to become cooler than normal which

increases the already abnormally high viscosity of the fluid. An increase in washing

time of the centrifugals is needed to get the required quality of sugar, increasing total

centrifuging and purging time. Also, the needle-shaped crystal reduces the purging

efficiency of the massecuites in the centrifugals resulting in a poor separation of

crystal and molasses, hence reducing the refined quality of the sugar. Additionally,

dust production was mainly due to the breaking and crushing of conglomerates and

needle crystals near to the screens of centrifugals. Consequently, sugar fine corn in

the final product is increased. Such micro-particles may be seen on the surface of a

sugar crystal in Figure 55. The conglomerates contain higher ash than

unconglomerated sugar. The higher ash is contained in trapped mother liquor, either

partially dried in the re-entrant angles, or in the inclusions in those angles (Madsen,

1980; van der Poel, 1980). In addition, our results (not view) elucidated increase of

the turbidity in sugar solution, which was produced from deteriorated beet syrup.


81

Figure 56 shows the effect of crystallization temperature on the surface topography

in presence of dextran at different temperatures. It is obvious that at a crystallization

temperature of 70°C more perfect crystal surfaces are built than at 55°C. As a

possible reason, the increase of viscosity caused by dextran and the effects of

temperature on the viscosity of (dextran-) sucrose solutions decribed in chapter 4.7

can be suggested. Also in crystal topography, the influence of dextran seems to be

diminished by the application of higher temperatures as it could be observed for

crystal growth rates.

This effect can also be correlated to the increase of colorant and turbidity

components of sugar produced at low temperatures (after product). Additionally, the

presence of dextran can assist the adsorption of these particles on the crystal surface.

6000x

6000x

55°C

70°C

55°C

70°C

Figure 56: Effect of crystallization temperature on the surface topography in the presence of dextran

Rogé et al., (2007) reported that there is a good correlation between white sugar

turbidity and calcium ion concentration in stored syrups. As well they reported that

macromolecular organic substances (as for example dextran) may act a carrier for

components causing turbidity.


82

In many factories, despite the use of ion exchange to reduce calcium ion, turbidity

phenomena in the produced sugar were observed. Figure 57 shows the comparison

between a single sugar crystal without turbidity and a sample of a crystal with the

highest amount of turbidity. Figure 58 is the magnification image of Figure 57,

which shows the non-sugar particles on the crystal surface. Hence, the biggest

particles (> 5µm) were calcium carbonate while the small particles (<5µm) were

calcium oxalate. These particles were formed during the storage of syrup under

adverse conditions.

a

b

a

b

Figure 57: Turbidity phenomena in many European sugar beet factories (a, b) during the thick juice campaign


83

Figure 58: The magnification image of Figure 57

In addition, an increase of elongated crystals in the after product crystallization step

as shown in Figure 59 was observed. These results are in good agreement with Rogé

et al., (2007), they reported that the elongation of face c is significant and such a

phenomenon increases with syrup turbidity.

Figure 59: After product sugar crystal in one of European sugar beet factory during the thick juice campaign

From the technical sucrose crystallization point of view, the three stage

crystallization scheme is regarded as the standard scheme. The process of producing

sugar is through several stages, because the high concentration of non-sugar and the


84

increase of solution viscosity hinder the sugar separation from the mother liquor.

Generally, in the most systems of beet sugar crystallization, the after product sugars

(C- sugar) are dissolved and recycled to the first crystallization stage, where they are

added to the thick juice (Figure 60). In addition, in other systems, the C-sugar is sent

to the second stage. This means, if sugar syrup contains a considerable amount of

dextran, the dextran levels in after product sugar will be increased, consequently, the

dextran levels in the previous crystallization stage increase continuously.

From the abovementioned results (comp. chapter 4.4, Figure 33 and Figure 34), 17-

20 % of the dextran remains in the after product. So, in this case, the amount of the

dextran, which is recycled to the first crystallization stage, can be calculated. This is

important as this dextran increases the original dextran level in thick juice (syrup).

For example, if dextran level in syrup is 1000 mg/kg DS, the dextran level in the

after product will be about 170 to 200 mg/kg DS. Consequently, the dextran level in

syrup becomes about 1170-1200 mg/kg DS. However, this increase can only be

characterized qualitatively in this work, as the remaining dextran depends for

example on the number of crystallization stages and on the industrial period. Also,

the relation between dextran in syrup and dextran in sugar (comp. Figure 33)

depends on parameters of centrifugation process.


85

1

0

30101

0

30401

0

3070

10

3050

10

3090

9

3055

3080

Melasse18,99 t/h

60,00% PurAIR/

VA POR

1

AIR

0

3100

Weiß-zucker

Roh-zucker

Nach-produkt

NachproduktKristalisator

190,01 t/h80,0 °C

RR

Weißzucker

151,94 t/h70,0 °C

63,94 t/h75,0 °C

WeißzuckerZentrifuge

RohzuckerZentrifuge

NachproduktZentrifuge

Dicksaft

Nachprodukt Kläre

Rohzucker Kläre

33,75 t/h80,1 °C

R

R

9

300

0

75,49 t/h82,0 °C

38,60 t/h87,0 °C

35,24 t/h10,49 t/h

Brüden 5

9

363

0

3910-0

Kondensation Kondensation Kondensation

R

3620-9

3770-0 4334-0 4344-0

52,05 t/h87,77 kPa

96,4 °C

0 8

9 3870

R4074-0

77,37 t/h99,99% Purity

3740-7

41 84-0

38 70 -6

41 74-0

387 0-3

NachproduktKläre

RohzuckerKläre

10

4054

1

0

4064

R

4264 -8

R

4054-0

R

1

0

4174

R228-7

1

0

4184

R2 28 -9

Dünnsaft

4194

Kondensat

Kondensat

1 0

0 8

9 4264

R

R42 64 -10

R3630-9

1 0

0 8

9 4284

R3630-7

AffinationAblauf

R4284 -1 0

4,42 t/h

10

4294

Kondensat

9

430

4

10

4314

R

3870-5

43 14-0Staub

Kondensat

3740-13740-3 3740-5

Dünnsaft

1

0

30101

0

30401

0

3070

10

3050

10

3090

9

3055

3080

Melasse18,99 t/h

60,00% PurAIR/

VA POR

1

AIR

0

3100

Weiß-zucker

Roh-zucker

Nach-produkt

NachproduktKristalisator

190,01 t/h80,0 °C

RR

Weißzucker

151,94 t/h70,0 °C

63,94 t/h75,0 °C

WeißzuckerZentrifuge

RohzuckerZentrifuge

NachproduktZentrifuge

Dicksaft

Nachprodukt Kläre

Rohzucker Kläre

33,75 t/h80,1 °C

R

R

9

300

0

75,49 t/h82,0 °C

38,60 t/h87,0 °C

35,24 t/h10,49 t/h

Brüden 5

9

363

0

3910-0

Kondensation Kondensation Kondensation

R

3620-9

3770-0 4334-0 4344-0

52,05 t/h87,77 kPa

96,4 °C

0 8

9 3870

R4074-0

77,37 t/h99,99% Purity

3740-7

41 84-0

38 70 -6

41 74-0

387 0-3

NachproduktKläre

RohzuckerKläre

10

4054

1

0

4064

R

4264 -8

R

4054-0

R

1

0

4174

R228-7

1

0

4184

R2 28 -9

Dünnsaft

4194

Kondensat

Kondensat

1 0

0 8

9 4264

R

R42 64 -10

R3630-9

1 0

0 8

9 4284

R3630-7

AffinationAblauf

R4284 -1 0

4,42 t/h

10

4294

Kondensat

9

430

4

10

4314

R

3870-5

43 14-0Staub

Kondensat

3740-13740-3 3740-5

Dünnsaft

Dust

Thick juice

White sugar centrifuge

Vapor 5

Molasses

Raw sugar centrifuge

Thin juice

White sugar

After product centrifuge

After product crystallizer

CondensationCondensationCondensation

Condensate

Condensate

Condensate

W hite sugar

Raw sugar

After product

Dissolved Raw sugar

Thin juice

Dissolved After product

Dissolved Raw sugarDissolved After product

Condensate

Affinatio run-off

Figure 60: 3-product scheme of crystallisation


86

4.9 Technical and technological consequences and future perspectives

The crystallization process depends on many factors, e.g. temperature,

supersaturation, solution viscosity, quantity and nature of the impurities. Dextran has

a pronounced effect on the important parameters influencing the crystallization rate

of sucrose solutions. Moreover, according to the achieved results it can be stated that

the presence of dextran molecules in sugar syrup has a strong effect on the

production and the quality of produced sugar in sugar factories. Figure 61

summarizes the most important parameters influencing the crystallization in sucrose

solutions. The solution purity (q), the non-sugar content (NS), the viscosity (η) and

the rate of diffusion (kD) were affected directly by the presence of dextran. The

above described results confirmed that the crystallisation rate at a constant

temperature is decreasing with higher dextran concentrations. At technologically

relevant supersaturations and temperatures, the decrease of crystallization rate,

depending on molecular weight of added dextran fractions, amounts up to 50 %.

q

DS

η

Stirring

Gro

wth

rate

of

sucr

ose c

rysta

lsTe

mpe

ratu

re

Rate

of

diffu

sion

Supe

r-Sa

tura

tion

Sucr

ose-

Solu

bilit

y

NS

Exhaustion of mother liquor

Relative velocity between crystal and mother syrup

Dextran molecules

Shape of crystalsNucleation rate

Figure 61: Influence of dextran on many crystallization parameters during crystallization

process.

Where: q = purity, NS = non-sugar, DS = dry substance and η = viscosity

Time reduction and low energy consumption in the crystallization process are of

particular interest in sugar factories. Figure 62 shows time periods for the reduction

of dry mass content by one percent for different dextran concentrations at 60°C and

70°C, respectively. At 60°C the crystallization periods were increased by 20 % and


87

68 % after the addition of dextran T2000 at concentrations of 1500 and 5000 mg/kg

DS, respectively. On the other hand, at 70°C the increase of crystallization time was

10 % and 21 %. Furthermore, an increase of crystallization temperature from 60°C to

70°C reduces the crystallization time by 8 %, 16 % and 33 % regarding admixtures

of dextran of 0 ppm, 1500 ppm or 5000 ppm, respectively. Moreover, the influence

of dextran increases strongly with both, the increase of molecular weight and the

decrease of crystallization temperature. Therefore, it is obvious, that a decrease in

crystallization temperature, e.g. from 70°C to 65°C, which is desirable regarding

energy aspects, is not applicable in occurrence of dextran. Technological measures to

manipulate only the high molecular fractions of dextran are not known so far.

0

0.5

1

1.5

2

2.5

Crys

talli

zatio

n tim

e (h

ours

)

0 mg/kg DS 1500 mg/kg DS 5000 mg/kg DS

Dextran content (mg/kg DS)

60°C 70°C

20%

10%

68%

21%

Figure 62: The relationship between dextran additions and crystallization time during the

crystallization process

In order to reach similar crystallization rates at the occurrence of dextran higher

supersaturations can be applied. However, the increase of supersaturation above a

certain level may cause spontaneous crystallization and thus the formation of fine

crystals. In practice, sugar factories have to work in the so called metastable zone

from 1 to 1.2 supersaturation (Brennan et al., 1976; van der Poel et al., 1998). Thus,

narrow boundaries are set considering an increase of supersaturation.


88

Basing on the abovementioned results future research topics are suggested on the

following topics:

Study on the effect of other organic non-sugar substances (e.g. other

macromolecular polysaccharide fractions or proteins) on sugar process

productions and sugar quality. Rogé et al. (2007) reported that organic

macromolecules act as carrier for non-sugar substances causing problems

during crystallization and causing turbidity in the final sugar. This effect was

also proved for dextran in this work. However, no comprehensive work is

known regarding the composition of organic macromolecules occurring

during sugar production.

In chapter 4.3 and 4.8 the relation between dextran concentration in syrup and

dextran concentration in sugar was illustrated. It was mentioned that this

relation is dependent on the crystallization and centrifugation regime. In order

to allow a precise process control to minimize the negative impact of dextran

a systematic investigation about the influence of relevant process parameters

(e.g. number of crystallization stages, amount of water for affination, type of

sugar production factory) on the transfer of dextran into the sugar has to be

realized. A qualitative and quantitative evaluation by determination of

dextran levels in the after product is suggested. This could be a valuable tool

to supplement available model based process control tools in the sugar

industry (Fiedler, et. al. 2007).

Based on the described effects of dextran on the crystal surface (s. chapter

4.8.2), future studies will investigate the relationship between dextran levels

and especially the drying process in order to characterize parameters

influencing the formation of micro-crystals and the incorporation of micro-

particles (colorants and turbidity) on the crystal surface.

Improvement of crystallization process at low temperatures (comp. chapter

4.6) by decreasing surface tension and viscosity using chemical agents during

the crystallization process.

Usage of enzyme (dextranase) during the extraction process of sugar beet

juices to improve the filtration process (Clarke,1997). The latter leads to an

increase of the soluble dextran amount (low molecular mass) in juice. This

means an increase of dextran levels in syrup and, hence, an increase of low

molecular weight dextran. The low molecular weight fraction of dextran is


89

known to affect the crystallization process and the sugar loss to molasses

(needle crystals), which was proved in the present study (see chapter 4.8.1).

Furthermore, the enzyme treatment leads to increase of glucose, which causes

an increase of color formation during the evaporation process (Maillard

reaction) (Kroh, et. al. 2007). In addition, it causes a reduction of

crystallization rate. The future work will focus on dextran molecular weight

fractions and sugars deriving from enzyme treatment and on the effect of

these products on the sugar production process.

Summary

90

5 Summary

In the present study, the sensitivity and accuracy of the common methods of dextran

determination were investigated. Concentrations and progressions of dextran during

different sugar production processes and industrial periods were determined and for

the first time an evaluation of sugar and molasses quality was provided based on the

measurement results. Regarding the dextran determination, the results had shown a

significant effect of dextran concentration and molecular fractions on the Haze and

Robert’s methods. The data demonstrate that the Robert’s method was more accurate

for different dextran molecular mass fractions than Haze method. Furthermore, the

Haze method was too inaccurate to be genuinely useful as a dextran analysis method

especially considering low dextran molecular mass fractions.

The isolation and identification of Leuconostoc mesenteroides as an important source

of dextran in sugar beet factories was performed. It was observed, that the

microorganisms grow during the storage of sugar beet and also during the extraction

process. The mean counts of colonies were reduced after washing process from

2.5*103 cfu/g of the sugar beet sample to 6*102 cfu/g of cossettes samples. However,

the number of microorganisms increased in raw juice after extraction process to

reach almost 4*106. In addition, these numbers depend on the degree of deterioration

of the sugar beets. Dextran content increased during the extraction process by 4 % to

20 % for fresh and deteriorated sugar beet, respectively. In contrast, the lime-carbon

dioxide juice purification reduced the dextran content in raw juice by 40 % to 60 %.

The obtained results from Egyptian sugar beet factories showed a correlation

between the increase of sugar loss and the presence of dextran during different

industrial periods as well as during the production process itself..

As a main focus in this work the negative effects of the presence of dextran during

(isothermal) crystallization were investigated. The growth rate of sucrose crystals

and the quality of the sugar production in pure sucrose solution at different

temperatures were studied. To elucidate the influence of dextran on the growth rate

of sucrose crystals, different molecular mass fractions of dextran from 40,000 g/mol

(T40) to 2,000,000 g/mol (T2000) were admixed in different concentrations to pure

sucrose solution. The most pronounced effect of dextran was found with T2000 at

Summary

91

60°C. On the other hand, negligible influences of dextran T40 and T500 were

observed at the same temperature and supersaturation. The high adsorption ability

and the increase of solution viscosity particularly caused by high molecular fractions

of dextran were identified as the main reasons for the reduction of crystal growth

rate.

The application of lower temperatures during crystallization enhances the problems

deriving from dextran. Regarding the final crystallization steps (e.g. after product

crystallization), levels of dextrans and impurities are enhanced and the mentioned

problems will be amplified in these production steps.

From the morphological studies, it was found that the presence of dextran in sugar

mother liquor leads to elongation of the sucrose crystal shape (elongated along the c-

axis). In particular, the lower molecular mass fractions of dextran are involved in this

effect. A scanning electron microscope (SEM) was utilized for the evaluation of

crystal surfaces and crystal morphology. The surface topography of sucrose crystals

was affected by the presence of higher dextran molecular mass fractions (T2000),

which causes gaps, rough areas and strong adsorption of micro-particles (micro-

crystals) on the crystal surface. Again, the described influences are enhanced by the

application of lower temperatures.

In addition, in this work it could be confirmed that there is a correlation between the

dextran content in syrup and quality of sugar produced, in particular the turbidity of

the sugar. So, a relation which has already been described in recent literature for

organic macromolecules in general could be specified for dextran molecular

fractions. .

Summary

92

Zusammenfassung

In dieser Studie wurde die Empfindlichkeit und Genauigkeit der gebräuchlichen

Methoden zur Dextranbestimmung untersucht. Basierend auf den in

unterschiedlichen Produktionsprozessen und industriellen Produktionsphasen

ermittelten Dextrangehalten wurde erstmals eine umfassende Bewertung des

Einflusses von Dextran auf die Zucker- und Melassequalität erstellt. Bezüglich der

Dextranbestimmung zeigten die Ergebnisse einen deutlichen Einfluss der

Dextrankonzentration und der molekularen Fraktionen auf die Haze- und Roberts-

Methode. Die Daten ergaben, dass die Robertsmethode für unterschiedliche Dextran-

Molekulargewichtsfraktionen eine größere Genauigkeit zeigt als die Haze-Methode.

Vor allem bei kleineren Dextran-Molekulargewichtsfraktionen war die Haze-

Methode nicht für genaue Bestimmungen geeignet.

Im Rahmen dieser Arbeit wurde Leuconostoc mesenteroides , eine der wichtigsten

Quellen von Dextran in Rübenzuckerfabriken isoliert und identifiziert. Diese

Mikroorganismen wachsen sowohl während der Lagerung der Zuckerrüben als auch

während des Extraktionsprozesses. Während des Waschens und Schneidens der

Rüben nahm die durchschnittliche Koloniezahl von 2.5*103 cfu/g in der Rübenprobe

bis auf 6*102 cfu/g in den Schnitzeln ab. Die Koloniezahl erreichte jedoch im Laufe

der Extraktion bis zum Rohsaft wieder einen Wert von 4*106. Die Zahlen waren

jedoch stark vom Frischegrad der verarbeiteten Rüben abhängig.

Die Konzentration an Dextran nahm während der Extraktion von frischen bzw.

verdorbenen Rüben um 4 bzw. 20 % zu. Im Gegensatz hierzu verringerte sich der

Dextrangehalt um 40 bis 60 % während der Saftreinigung mit Kalk und CO2.

Ergebnisse aus ägyptischen Rübenzuckerfabriken zeigen einen klaren

Zusammenhang zwischen der Zunahme des Zuckerverlustes und der

vorherrschenden Dextrankonzentration während verschiedenen Perioden der

Zuckerkampagne aber auch während den einzelnen Produktionsabschnitten auf.

Einen Schwerpunkt dieser Arbeit stellt die Untersuchung der negativen Einflüsse von

Dextran während des (isothermen) Kristallisationsprozesses aus reinen

Saccharoselösungen bei verschiedenen Temperaturen dar. Dextranfraktionen

verschiedener Molekularmassen zwischen 40000 g/mol (T40) und 2000000 g/mol

Summary

93

(T2000) wurden reinen Saccharoselösungen in verschiedenen Konzentrationen

zugesetzt, um den Einfluss von Dextran auf die Kristallwachstumsrate genau

charakterisieren zu können. Der stärkste Einfluss war bei dem Zusatz der Fraktion

T2000 bei 60°C zu detektieren. Im Gegensatz hierzu besaßen die Dextranfraktionen

T40 und T500 einen wesentlich geringeren Einfluss auf die Wachstumsrate bei

jeweils gleichen Temperaturen und Übersättigungen. Gründe hierfür liegen in der

hohen Adsorptionsfähigkeit und der Erhöhung der Viskosität der Zuckerlösung

insbesondere durch die hochmolekulare Fraktion von Dextran.

Die Absenkung der Temperaturen während der Kristallisation, wie es aus

ökonomischen Gründen häufig angestrebt wird, vergrößert die durch Dextran

ausgelösten Probleme. Insbesondere in letzten Kristallisationsstufen (z.B.

Nachproduktkristallisation) treten erhöhte Konzentrationen an Dextran und Nicht-

Zuckerstoffen auf, wodurch die beschriebenen Effekte noch verstärkt werden.

Um den Einfluss von Dextran auf die Kristallmorphologie und die Kristalloberfläche

evaluieren zu können, wurden elektronenmikroskopische Aufnahmen angefertigt.

Die Ergebnisse zeigen, dass die Kristallform (Verlängerung in Richtung der c-Achse)

durch die niedermolekulare Fraktion (T40) beeinflusst wird. Demgegenüber hat die

hochmolekulare Dextranfraktion T2000 einen starken Einfluss auf die Qualität der

Kristalloberfläche. Beobachtet wurden raue, lückenhafte Kristalloberflächen und die

Anhaftung von Mikro-Partikeln (Mikro-Kristalle) auf der Oberfläche. Die

beschriebenen Effekte werden wiederum durch den Einsatz tiefer Temperaturen

verstärkt.

In dieser Arbeit konnte ebenfalls ein Zusammenhang zwischen dem Dextrangehalt

und der Qualität des Zuckers, insbesondere der Zuckertrübung, gezeigt werden. Dies

spezifiziert für Dextran einen Zusammenhang, der in der aktuellen Literatur

allgemein für organische Makromoleküle schon beschrieben wird.

References

94

6 References

Alsop, R. M. (1983): Industrial Production of Dextrans. Progress in Industrial Microbiology. 1-42. ed. M. E. Bushell. New York: Elsevier.

AOAC. (1990): Official Methods of Analysis, Arlington, VA.

Atkins, P. C. and McCowage, R. J. (1984): Dextran - A review. The Australian experience proceeding of the International Dextran Workshop, New Orleans, Louisiana, USA, 7-39.

Austmeyer, K. E. (1981): Untersuchungen zum Wärme-und Stoffübergang im Anfangsstadium der Verdampfungskristallisation der Saccharose, Techn. Univ. Carlo-Wilhelmina zu Braunschweig. Germany

Back, W. (1994): Kultivierung/Methoden, Brauerei, Winzerei, Bollmann Druck GmbH, Zirndorf, Germany.

Battcock, M. and Ali, S. A. (1998): Bacterial fermentations. Fermented fruits and vegetables a global perspective, V. d. T. d. C. FAO, ed., Food and Agriculture Organization, Rome, Italy.

Beukes, E., M., Bester, B., H. and Mostert, J. F. (2001): The microbiology of South African traditional fermented milks. International Journal of Food Microbiology, 63, 189-197.

Bose, S. and Singh, L. (1981): Problems associated with dextran, the major microbiological cause of sucrose loss and their remedies. Indian Sugar, December, 603-608.

Brennan, J. G., Butters, J. R., Cowell, N. D. and Lilly, A. E. V. (1976): Food Engineering Operations, Applied Science Publishers, London.

Brown, C. F. and Inkerman, P. A. (1992): Specific method for quantitative measurement of the total dextran content of raw sugar. J. Agric. Food Chem., 42, 227-233.

Bubnik, Z. and Kadlec, P. (1996): Formulas for calculation of properties of sugar solutions and sucrose crystals. Zuckerindustrie, 121(1), 36-39.

Bubnik, Z. Vaccari, G., Mantovani, G., Sguadino, G., and Kadlec, P. (1992): Effect of dextran, glucose and fructose on sucrose crystal elongation and morphology. Zuckerindustrie, 117(7), 557-561.

Bunert, U., Bruhns, M. (1995): Thermische Bedingungen für die Silolagerung von Weißzucker und Trockenschnitzelpellets. Zuckerindustrie 120 (4), 265-272.

Cbristodoulou, P. (2000): New approaches to crystallization theories. Abstracts for SPRI 2000 conference on sugar processing research, Porto, Portugal.

Chung, C. C. (2000): A manual for the design and operation of sugar refining facilities. Handbook of sugar refining, John Wiley & Sons, New York., 516-517.

Clarke, M. A., Edye, L. A., Cole, F. and Kitchar, J. L. (1997): Sugarcane factory trials with dextranase enzyme. Sugar Journal, 60, 20-22.

References

95

Clarke, M. A. and Godshall, M. A. (1988): Determination of dextran in raw cane sugar by Robert's copper method: Collaborative study. Journal Association of Analytical Chemists, 71(2), 276-279.

Clarke, M. A., Roberts, T. B. T. and Tsang, W. S. C. (1986): Update on dextran and dextran analysis. Presented at S. I. T, Publication of Technical Papers and Proceedings of the Sixty-Third Annual Meeting of Sugar Technologists, Vancouver, Canada, paper No. 544.

Cossairt, G. W. (1982): Crystallization. Beet-Sugar Technology, R. A. McGinnis, ed., Beet Sugar Development Foundation, USA, 371-465.

Cuddihy, J. A. and Day, D. F. (1999): The process and financial impact of dextran on sugar factory. Midland Research Laboratories, Inc.

Cuddihy, J. A., Miguel, E. P. and James, S. R. (2000): The presence of total polysaccharides in sugar production and methods for reducing their negative effects. Midland Research Laboratories, Inc.

Curtin, J. H. and McCowage, R. J. (1986): Dextran measurement in cane products. 19th Congress ISSCT, 755-764.

Day, D., Cuddihy, J. A. and Rauh, J. S. (2002): Versatility of the Antibody dextran test method. Midland Research Laboratories, Lenexa, Kansas, USA.

Day, D. F. (1984): Dextran control in the sugar house. Sugar Journal, 46, 16-18.

Day, D. F. (1986): Dextran-problems, solutions, current research. Paper presented at the annual meeting of the American Society of Sugar Cane Technologists, Baton Rouge, LA, USA, 6-7 Feb.

Day, D. F. (1992): Spoilage in the sugar industry. The lactic acid bacteria, B. J. B. Wood, ed., Elsevier Applied Science, London and New York., 343-361.

De Man, J. C., Rogosa, M. and Sharpe, M. E. (1960): A medium for the cultivation of lactobacilli. J. Appl. Bacteriol., 23, 130-135.

DeStefano, R. P. and Irey, M. S. (1986): Measuring dextran in raw sugars historical perspective and state of the art. Journal American Society Sugar Cane Technologists, 6, 112-120.

Edye, L. A., Clarke, M. A., Roberts, E. J. and Blanco., R. S. (1995): Dextran analysis by N.M.R. Publication of Technical Papers and Proceedings of the Sixty-Third Annual Meeting of Sugar Technologists, Vancouver, Canada, S. I. T, 55-64.

Eggleston, G. and Monge, A. (2005): Optimization of sugarcane factory application of commercial dextranases. Process Biochemistry, 40(5), 1881-1894.

Ekelhof, B. (1997): Kristallisationskinetik der Saccharose in reinen und unreinen Lösِungen. Dissertation, Techn. Univ. Braunschweig, Germany.

Elahi, M. (2004): Untersuchung zur Optimierung der Kühlrate der Kühlungskristallisation von Nachprodukt-Kristallsuspensionen bei der Saccarosegewinnung. Dissertation, Techn. Univ. Berlin, Germany.

Elwell, D. and Scheel, H. J. (1975): High temperature solutions, Academic Press, London.

References

96

Fiedler, Th., Smejkal, Q., Kroh, L.W. and Kurz T. (2007): Simplified Kinetics and Colour Formation in Sucrose Solutions Based on a-Dicarbonyl Compounds. Int J Food Engineering, submitted

Frenkel. (1958): Technological of beet sugar and raffination industry Moscou.

Galea, C. F. and Inkerman, P. A. (1993): Dextran analysis of raw sugar. Part I. A specific method for total dextran. International Sugar Journal, 95 (1136), 309-313.

Gillet, E. C. (1977): Crystallization in Motion. Cane Sugar Handbook, Meade and Chen, eds., John Wiley & Sons. Inc., Canada, 290-333.

Godshall, M. A., Legendre, B. L., Clarke, M. A., Miranda, X. M. and Blanco, R. S. (1994): Starch, polysaccharide and proanthocyanidin in Louisiana sugarcane varieties. International Sugar Journal, 98, 144-148.

Golovin, P. V. and Grerasimenko, A. A. (1959): Crystallization velocity of sucrose at higher temperatures. ISSCT, 248-262.

Grimsey, I. M. and Herrington, T. M. (1996): The incorporation of colored compounds in sucrose crystals. Zuckerindustrie, 121, 40-45.

Gudoshnikov, S. (2007): World sugar market-Back to Surplus. World Perspectives for Sugar Beet and Cane as a Food and Energy Crop, Sharm El Sheikh, GL1.1/1-11.

Gudshall, M. A. (1992): The isolation of a high molecular weight colorant from white beet sugar. Sugar Processing Research, S. P. R. I, Inc., 312-319.

Guglielmone, C. d., Diez, O., Cardenas, G. and Oliver, G. (2000): Sucrose utilization and dextran production by Leuconostoc mesenteroides isolated from the sugar industry. Sugar Journal, 62, 36-40.

Gumaraes, L., Susana, S., Bento, L. S. M. and Rocha, F. (1995): Investigation of crystal growth in a laboratory fluidized bed. International Sugar Journal, 97, 199-204.

Gupta, A. P. and Prabhu, K. A. (1993): Control of Leuconostoc contamination in sugar industry. International Sugar Journal, 95(11379), 93-97.

Halebian, S., Harris, B., Finegold, S. M. and R D Rolfe, R. D. (1981): Rapid method that aids in distinguishing Gram-positive from Gram-negative anaerobic bacteria. J. Clin. Microbiol., 13(3), 444-448.

Hamasaki, Y., Ayaki, M., Fuchu, H., Sugiyama, M. and Morita, H. (2003): Behavior of Psychrotrophic Lactic Acid Bacteria Isolated from Spoiling Cooked Meat Products. Appl Environ Microbiol, 69(6), 3668-3671.

Hartman, P. and Perdok, W. G. (1955): On the relations between structure and morphology of crystals I. Acta Crystallographica, 8, 49-52.

Haynes, L. (2004): Dextran in Refined Sugar. 2004 SPRI Conference Abstracts of Presentations, April, 4-7, 2004.

Hehre, E. J. and Sugg, J. Y. (1942): Serologically reactive polysaccharides produced through the action of bacterial enzymes: I. Journal of Experimental Medicine, 75, 339-353.

References

97

Holzapfel, W. and Kandler, O. (1969): Zur Taxonomie der Gattung Lactobacillus Beijerinck. VI. Lactobacillus coprophilus subsp. confusus nov. subsp., eine neue Unterart der Untergattung Betabacterium, Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt., 2(123), 657-666

Holzapfel, W. and Schillinger, U. (1992): The Genus Leuconostoc. A. Barlows; H. G. Trüper;M. Dworkin;W. Harder and K. H. Schleifer, eds., the Prokaryotes, Berlin.

Houghton, B. J., Dutton, J. V., Harvey, C. W. and Sissons, A. (1998): Sugar Technology Beet and Cane sugar manufacture, P. W. v. d. Poel; H. Schweck and T. Schwartz, eds., Beet Sugar Development Foundation, Denver, USA, 653-656.

Hucker, G. J. and Pederson, C. S. (1930): Studies on the Coccaceae. XVI. The genus Leuconostoc, New York State Agricultural Experiment Station, 167, 3-8.

ICUMSA. (1994): International commission for uniform methods of sugar analysis. Colney, England. GS 1-15.

Imrie, F. K. E. and Tilbury, R. (1972): Polysaccharides in sugar cane and its products. Sugar technol. Rev., 1, 291-361.

Inkerman, P. A. (1980): An appraisal of the use of dextranase. 17th Congress Int. Soc. Sugar Cane Technol, Manila, Philippines, 2411–2427.

James, A. C. and Day, D. F. (2000). The process and financial impact of dextran on a sugar factory. Midland Research Laboratories, Inc.

James, A. C., Miguel E. P. and James. S. R. (2000). The presence of total polysaccharides in sugar production and methods for reducing their negative effects. Midland Research laboratories, Inc.

Jeanes, A., Haynes, W. C., Wilham, C. A., Rankin, J. C., Melvin, E. H., Austin, M. J., Cluskey, J. E., Fisher, B. E., Tsuchiya, H. M. and Rist, C. E. (1954): Characterization and classification of dextrans from ninety-six strains of bacteria. Journal of the American Chemical Society, 76, 5041-5052.

Jimenez, E. R. (2005): The dextranase along sugar-making industry. Biotecnologia Aplicada, 22(1), 20-27.

Karagodin, M. (1998): New batch centrifugal performance at a raw sugar factory. Sugar y Azucar, Oct., 26-38.

Kennedy, J. F. and White, C. A. (1988): Carbohydrate Chemistry, Oxford University, New York.

Kenter, C. and Hoffmann, C. (2006): Changes in the quality of frost-damaged sugarbeet during storage as affected by temperature and variety. Zuckerindustrie, 13(2), 85-91.

Khaleifah, M. A. (2001): Effect of Dextran on Sugarcane quality and raw sugar manufacture. Dissertation, Assiut univ., Egypt.

Kim, D. and Day, D. F. (2004): Detrmination of dextran in raw sugar process streams. Food Science and Biotechnology, 13(2), 248-252.

Kobayashi, M. and Matsuda, K. (1974): The dextransucrase isoen-zyme from L. mesenteroides NRRL B-512F. Biochim. Biophys. Acta, 370, 441-449.

References

98

Kroh, L. W. Fiedler, T. and Wagner, J. (2007): The formation of colored compounds in technical sucrose solutions – new knowledge about their molecular basics and molecular size distribution. Zuckerindustrie 132 (1), 25–32.

Kruse, M. and Ulrich, J. (1993): Kristallwachstum in der Lösugskristallisation als Drei-Stufen-Modell, Chem. Ing. Techn., 65, 60-63.

Leathers, T. D., Hayman, G. T. and Cote, G. L. (1995): Rapid Screening of Leuconostoc mesenteroides Mutants for Elevated Proportions of Alternan to Dextran. Current Microbiol, 31, 19-22.

Lonvaud-Funel, A. (1999). "Leuconostoc." Academic Press, University Victor Segalen, Bordeaux, France.

Madsen, R. F. (1980): Rührwerke in Kochapparaten. Erfahrungen der DDS. Zuckerindustrie, 105, 234-237.

Mantovani, G., Gill, I. G. and Fagioli, F. (1967): Einfluss der Kristallstruktur von Nichtzuckerstoffen auf die Änderung des Habitus von Saccharose. Zucker, 20, 663-668.

Mantovani, G. and Vaccari, G. (1998): Crystal growth. Sugar Technology, P. W. van der Poel; H. Schiweck and T. Schwartz, Bartens, Berlin, 653-656.

Martins, P. and Rocha, F. (2000): Using Volumetric Growth Rates to Characterize Sucrose Crystallization. Abstracts for SPRI 2000 conference on sugar processing research, Porto, Portugal.

Maurandi, V., Mantovani, G., Vaccari, G. and Rossi, A. (1988): Kinetik und Technologie der Erschöpfung von Nachprodukt-Kochmassen. Zuckerindustrie, 113, 791-794.

McGinnis, R. A. (1982): Beet-Sugar Technology, Beet Sugar Development Foundation, USA.

Meade, G. P. and Chen, J. C. P. (1977): Cane Sugar Handbook, John Wiley and Sons, New York.

Miller, K. F. and Wright, P. G. (1977): Experiments with molasses exhaustion. The Queensland Society of Sugar Cane Technologists, Watson Ferguson and Company, Brisbane, Queensland, Australia, 205-213.

Nicholson, R. I. and Horsley, M. (1959): Impurities in Sugar Processing, Determination of Dextran and Starch in Cane Juices and Sugar Products. J Agric Food Chem, 7(9), 640-643.

Noyes, A. A. and Whitney, N. R. (1897): The Rate of solution of solid substances in their own solutions. Journal of American Chemical Society, 19, 930-934.

Parrisch, F. W. and Clarke, M. A. (1983): Effects of oligosaccharides and polysaccharides one sucrose crystallization. Tienen, 467-481.

Paton, N. H., Player, M. R. and Urquhart, R. M. (1994): A modified dextran assay for raw sugar, S.I.T., Publication of Technical Papers and Proceedings of the Sixty-Third Annual Meeting of Sugar Technologists, Vancouver, Canada, Paper 666.

References

99

Pot, A., Kuijvenhoven, L. J., Nie, d., L. H. and Jong, d., E. J. (1984): A Study on the Formation Factors of Crystal Size Distribution in Industrial Sucrose Crystallizers. Zuckerindustrie, 109, 305-313.

Rauh, J. S., Cuddihy, J. A. and Falgout, R. N. (1999): Analyzing dextran in the sugar industry: A review of dextran in the factory and a new analytical technique. Midland Research Laboratories.

Roberts, E. J. (1983): A quantitative method for dextran analysis. International Sugar Journal, 85, 10-13.

Robyt, J. F. (1986): Dextran. Encyclopedia of Polymer Science and Engineering, H. F. Mark; N. M. Bikales; C. G. Overberger and M. G., eds., John Wiley & Sons, New York, NY, 752-767.

Rogé, B., Bensouissi, A. and Mathlouthi, M. (2007): Effect of calcium on white sugar turbidity. Zuckerindustrie, 132(3), 170-174.

Sancho, D., Deban, L., Martinez, L., Pardo, R. and Vega, M. (1998): Direct determination of dextran in sugar juices from sugar refining processes by stripping voltammetry. Journal Environ. Sci. Health, A 33(6), 1171-1184.

Sarkar, D. and Day, D. F. (1986): Dextran analysis: A modified method. J. Amer. Soc. Sugar Cane Tech., 6, 102-107.

Sarkar, D., Day, D. F., Clarke, S. J. and Saska, M. (1991): Dextran analysis a comparison of methods. J. Amer. Soc. Sugar Cane Tech., 11, 69-99.

Saska, M. and Oubrahim, Y. (1987): Gel permeation chromatography of sugarcane products. Sugar Journal(June), 22-25.

Scheibler, C. (1874): Untersuchungen über die Natur der gallertartigen Ausscheidung (sog., Froschlaich), welche bei der Saftgewinnung aus Rüben beobachtet wird. Zeitschrift des Vereins der Rübenzucker Industrie, 24, 309-319.

Schiweck, H. and Mannheim, A. S. (1998): Batch evaporating crystallization. Sugar Technology, P. W. Van der Poel; H. Schiweck, and T. Schwartz, eds., Bartens, Berlin, 714-720.

Schliephake, D. and Austmeyer, K. E. (1976): Einige Aspekte zur Theorie und Praxis der Zuckerkristallisation. Zucker, 29, 293-301.

Schliephake, D., and Ekelhof, B. (1983): Beitrag zur vollständigen Berechnung der Kristallisationsgeschwindigkeit der Saccharose. Zuckerindustrie, 108(12), 1127-1137.

SCIMAT. (2006). "Leuconostocs " http://www.magma.ca/~pavel/science/ Leuconostoc.htm.

SIL. (2007). "International Sugar Stats." SUGAR ILLOVO LIMITED, http://www.illovosugar.com/worldofsugar/internationalSugarStats.htm.

Silin, P. M. (1958): La Technologie de Production de Sucre de Betterave et le Raffinage. Wyd. Piszezepromizdat Moscou.

Singleton, V. (2002): Review article: advances in techniques of dextran analysis - A modern day perspective. International Sugar Journal, 104(1239), 132-136.

http://www.illovosugar.com/worldofsugar/internationalSugarStats.htm

References

100

Singleton, V., Horn, J., Bucke, C. and Adlard, M. (2001): A new polarimetric method for the analysis of dextran and sucrose. International Sugar Journal, 103(1230), 251-254.

SPSS-Inc. (1998). "SPSS for windows, release 9, SPSS-Inc."

Steinbuechel, A. and Rhee, S. K. (2005): Polysaccharides and Polyamides in the food Industry, Wiley-VCH Verlag GmbH & CO. KGaA.

Südzucker, AG (2007): http://www.suedzucker.de/produkt/schema/index.shtml.

Sutherland, D. N. and Paton, N. (1969): Dextran and crystal elongation. International Sugar Journal, 71(131-135).

Sutherland, I. W. (1996): Extracellular Polysaccharides. Biotechnology, G. R. H.-J. Rehm, A. Puhler and P. Stadler, ed., Products of Primary Metabolism, New York: Wiley-VCH Verlag. GmbH & CO. KGaA.

Thelwall, C. (2002): Evaporative crystallisation. Publication of Technical Papers and Proceedings of the Sixty-Third Annual Meeting of Sugar Technologists, Vancouver, Canada.

UI-Qader, S. A., Iqbal, L., Rizvi, H. A. and Zuberi, R. (2001): Production of dextran from sucrose by a newly isolated strain of Leuconostoc mesenteroides (PCSIR-3) with reference to L. mesenteroides NRRL B-512F. Biotechnol. Appl. Biochem., 34, 93-97.

UL-Qader, S. A., Iqbal, L., Aman, A., Shireen, E. and Azhar, A. (2005): Production of Dextran by Newly Isolated Strains of Leuconostoc mesenteroides PCSIR-4 and PCSIR-9. Turkish Journal of Biochemistry, 31(1), 21-26.

Urquhart, R. M., McCowage, R. J. and Inkerman, P. A. (1993): Dextran analysis of raw sugar. part II. Acomparison of methods. International Sugar Journal, 95, 344-346.

Vaccari, G., Mantovani, G., Sguadino, G., Aquilano, D. and Rubbo, M. (1986): The raffinose effect on sucrose morphology and kinetics. Sugar technol.Rev, 13, 133-178.

Vaccari, G., Mantovani, G., Sguadino, G., Tamburini, E. and Aquilano, D. (1999): Fructo-oligosaccarides and Sucrose Crystal Growth Morphology. Zuckerindustrie, 124(34-39, 536-541).

Vaccari, G., Wawro, P., Tamburini, E., Sguadino, G. and Bernardi, T. (2002): Cooling crystallization of microfiltered raw juice and of Traditional thick juice: A Comparison. Zuckerindustrie, 127(22-28).

van der Poel, P. W. (1980): Qualitätsbeeinflussung der Zuckerkristalle durch die Arbeit mit Rührwerken Erfahrungen bei der CSM. Zuckerindustrie, 105, 237-240.

van der Poel, P. W., Schiweck, H. and Schwartz, T. (1998): Sugar Technology, Bartens, Berlin.

Vavrinecz, G. (1965): Atlas der Zuckerkristalle, Verlag Dr. Albert Bartens KG, Berlin.

References

101

Wagner, G. (2003): Entwicklung eines automatisierten Bildanalyseverfahrens zur Beurteilung der Kristallqualität in Saccarosekristallsuspensionen, Techn. Univ. Berlin, Germany.

Walstra, P. (2003): Physical Chemistry of foods, Marcel Dekker, Inc, New York, USA.

Wells, W. D. and James, G. P. (1976): Rapid dextran formation in stale cane and its processing consequences. Queensland Society of Sugar Cane technologists, 287-293.

Wilson, T. E. (1996): A comparison of raw sugar polarisation methods. International Sugar Journal, 98(1168), 169-174.

Wittenberg, A. (2001): Beitrag zur Optimierung der technischen Saccharosekristallisation. Dissertation, Techn. Univ. Berlin.

Yanli. Mi. (2002): Protection Mechanisms of Excipients on Lactate Dehydrogenase during Freeze-Thawing and Lyophilization. Dissertation, Tennessee., Memphis.

Appendix

102

7 Appendix

Appendix 1: The characterization morphology of isolated strains

Strains Source Isolation Colony morphology

Microscopy Notices

S1 Raw juice

Round, entire margin, white, convex, glistening, d=1mm

Coccus, single

Bacteria 7*102 cfu/g

S2 Raw juice

Round, entire margin, convex, glistening, centre structured/clear, margin translucent, d = 1 mm

Coccus, single


S3 Raw juice

Round, entire margin, centre structured/unclear, margin translucent

Coccus, Single and

diplococcus


S4 Raw juice

Like S1 single Yeast , 1*102 cfu/g

S5 Raw juice

Round, lobate, glistening, but lobate / unclear d= 1.5 mm

Coccus, Diplococcus


S6 Raw juice

Like S4 but translucent

single Yeast cells 8*102 cfu/g

S7 Raw juice

Round, translucent, filamentous, lobate,bristly, d = 3.5 mm

single Yeast 2*102 cfu/g

S8 Raw juice

Round, entire margin, convex, yellow-brown, dull, d=1.5 mm

Yeast cells 1*103 cfu/g

S9 Raw juice

white, entire margin, irregular, glistening, d=2 mm

Yeast cells 1*103 cfu/g

S10 Cossettes

Round, entire margin, red, convex, not translucent, d=2 mm

Yeast cells

S11 Cossettes

lobate, no entire margin

Yeast cells

S12 Raw juice

Like S1 Coccus Bacteria

S13 Raw juice


S14 Raw juice

Media: MRS-Agar

Temperature: 30°C


Appendix

103

Appendix 2: Characterizations of the isolated bacteria strains

Test Strains

Gas formation Gram (KOH-

Test)

Catalase test Acid

formation

S1 + + - +

S2 + + + +

S3 - + + +

S5 + + - +

S12 + + - +

S13 + + - +

S14 + + - +

Appendix

104

Appendix 3: Analyze the API 50 CHL Test for Leuconostoc

Sugar

Strains 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Leu.m.ssp cremoris - - - - - - - - - - +++ +++ - - - - - - - - - - + ++ - - - Leu.m.ssp m../ dextranicum1 - - - - +++ ++ +++ - - - +++ +++ +++ +++ - - - - - - - +++ +++ +++ +++ +++ Leu.m.ssp m./ dextranicum2

Time (Day)

- - - - +++ +++ +++ ++ +++ +++ +++ - - - - - - - +++ +++ - + ++

1 - - - - +++ ++ +++ - - - - +++ ++ + - - - - - - - +++ ++ ++ ++ 2 - - - - +++ ++ +++ - - - + +++ +++ +++ - - - - - - - +++ +++ +++ ++ +++ 5 - - - - +++ ++ +++ - - - +++ +++ +++ +++ - - - - - - - +++ +++ +++ +++ +++

S1

1 - - - - - +++ +++ - - - +++ +++ +++ +++ - - - - ++ - +++ - +++ - ++ ++ 2 - - - - - +++ +++ - - - +++ +++ +++ +++ - - - - +++ - +++ - +++ ++ +++ +++ 5 - - - - - +++ +++ - - - +++ +++ +++ +++ - - - - +++ - +++ - +++ +++ +++ +++

S5

1 - - - - +++ - - - - - - +++ ++ ++ - - - - - - - - ++ + + ++ 2 - - - - +++ + - - - - + +++ +++ +++ - - - - +++ - +++ ++ ++ ++ + +++ 5 - - - - +++ + - ++ +++ +++ +++ - - - - +++ - +++ ++ ++ ++ ++ +++

S12

1 - - - - +++ - - - - - - +++ ++ - - - - - - - - - ++ + + ++ 2 - - - - +++ + - - - - - +++ +++ ++ - - - - +++ - - ++ +++ +++ ++ +++ 5 - - - - +++ + - - - - - +++ +++ +++ - - - - +++ - - +++ +++ +++ ++ +++

S13

1 - - - - +++ - - - - - ++ +++ +++ +++ - - - - +++ - - - +++ ++ ++ ++ 2 - - - - +++ - - - - - +++ +++ +++ +++ - - - +++ - - - +++ +++ +++ +++

S14 5 - - - - +++ - - - - - +++ +++ +++ +++ - - - - +++ - - - +++ +++ +++ +++

Appendix

105

Sugar Strains 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

Leu.m.ssp cremoris - - - + - + - - - - - - - - - - - - - - - - - - Leu.m.ssp m../ dextranicum1 +++ +++ +++ ++ +++ +++ +++ - - ++ - - - +++ +++ - - - - - - + - - Leu.m.ssp m./ dextranicum2

Time (Day)

+ + +++ + +++ +++ +++ - - +++ - - - + +++ - - - - - - + - +

S1 1 - ++ +++ - +++ +++ - - - +++ - - - ++ +++ - - - - - - + - + 2 - +++ +++ - +++ +++ +++ - - +++ - - - ++ +++ - - - - - + - + 5 +++ +++ +++ ++ +++ +++ +++ - - +++ - - - ++ +++ - - - - - - + - + S5 1 ++ +++ +++ ++ ++ +++ +++ - - - - - - - ++ - - - - - - - - - 2 +++ +++ +++ +++ +++ +++ +++ - - - - - - - ++ - +++ - - - - - - - 5 +++ +++ +++ +++ +++ +++ +++ - - - - - - - ++ +++ - - - - - - - S12 1 ++ + +++ - +++ ++ +++ - - - - - - ++ ++ - - - - - - + - + 2 ++ +++ +++ - +++ +++ +++ - - +++ - - - ++ +++ - +++ - - - - ++ - ++ 5 ++ +++ +++ ++ +++ +++ +++ - - +++ - -- ++ +++ - +++ - - - - - - - S13 1 + - +++ - +++ +++ +++ - - - - - - - +++ - - - - - - + - - 2 ++ ++ +++ +++ +++ +++ +++ - - +++ - - - ++ +++ - +++ - - - - ++ - - 5 ++ +++ +++ +++ +++ +++ +++ - - +++ - - - ++ +++ - +++ - - - - ++ - - S14 1 ++ +++ +++ ++ - - +++ - - - - - - ++ - - - - - - - ++ - - 2 ++ +++ +++ ++ - - +++ - - - - - - ++ - - + - - - - ++ - - 5 ++ +++ +++ ++ +++ +++ ++ + +++ ++

C.V. and list of publications

106

8 C. V. and List of Publications

Curriculum Vitae / Lebenslauf

El-Sayed Ali Abdel-Rahman Soliman geboren am 20.04.1968 in Sohag (Ägypten)

Hochschulausbildung:

1987- 1991

Studium Lebensmitteltechnologie an der Assiut Universität, Ägypten, Abschluss Bachelor

1993 – 1999

Studium Lebensmitteltechnologie “Biochemical and Technological studies on Mungbean seeds” an der Assiut Universität, Ägypten, Abschluss Magister

1999-2003

Oberassistent an der Abteilung Lebensmitteltechnologie, Fakultät Landwirtschaft, Assiut Universität

ab 2003

Stipendium der ägyptischen Regierung für promovierte Wissenschaftler

ab Juli 2004

Wissenschaftlicher Mitarbeiter am Fachgebiet Lebensmittelverfahrenstechnik an der TU Berlin Promotionsthema “Untersuchungen zum Einfluss höhermolekularer Polysaccharidfraktionen auf das Kristallwachstum in Saccharoselösungen”

ab Juli 2006 bis Juli 2007

Stipendium von der TU-Berlin zum Thema “Untersuchungen zur Bildung und Beseitigung von Trübungen aus konzentrierten technischen Saccharoselösungen”

Berufstätigkeit: Praktikum

• Training course on modern techniques in molecular

biology and their practical applications (Assiut Uni.) • Praktikum in der Zuckerfabrik Jülich AG, Thema “

Untersuchung zur Trübsaftbildung im gelagerten Dicksaft” (Deutschland)

• Praktikum in den Zuckerfabriken Aarberg und Frauenfeld zum Thema “ Dar Zusammenhangs zwischen Saftqualität und Zuckerqualität in der Dicksaftkampagne ” (Schweiz)

• Praktikum am FG Lebensmittelverfahrenstechnik, TU-Berlin, zum Thema Zuckertechnologie (Deutschland)

• PIW Projekt Prozessingenieurwissenschaften im FG Lebensmitteverfahrenstechnik, TU-Berlin, zum Thema „Zuckerbäckerei“ (Deutschland)

Lehrerfahrung:

Agricultural Biochemistry, Principles of Food Technology, Food Chemistry and Analysis, Food Chemistry, Sugar Technology and Special Products Technology


107

List of Publications

Veröffentlichungen in Zusammenhang mit dieser Arbeit / Publication in

conjunction with this thesis

[1] E. A. Abdel-Rahmana; Q. Smejkal, R. Schick, El-Syiad, S. and Kurz, T. (2006)

Influence of high dextran concentrations and molecular fractions on the rate of

sucrose crystallization in pure sucrose solutions. Journal of Food Engineering,

Available online 5 July 2007

[2] E. A. Abdel-Rahman, R. Schick, T. Kurz (2007) Influence of dextran on sucrose

crystallization. Zuckerindustrie, 132 (6), 453-460.

Weitere Veröffentlichungen / Other Publications

[1] El-Rify, M.N.; El-Geddawy, M.A.H; El-Fishawy, F.A; and Abdel-Rahman, E.

A. (2000) Effect of processing on the amino acid composition and protein quality of

mung bean seeds. Mansoura University Journal of Agricultural Sciences, pp. 93-102.


A. (2000) Effect of some technological processes on the gross chemical composition

and mineral contents of mung bean seeds. Mansoura University Journal of

Agricultural Sciences, pp. 103-113.

[4] E. A. Abdel-Rahmana, M. A. El-Geddawyb, F. A. El-Fishawyb, T. Kurz and M.

N. El-Rify (2006) Isolation and physico-chemical characterization of mung bean

starches, International of Food Process Engineering (submitted).

[5] E. A. Abdel-Rahmana, M. A. El-Geddawyb, F. A. El-Fishawyb, T. Kurz and M.

N. El-Rify (2006) The changes in the lipid composition of mung bean seeds as

affected by processing methods. International of Food Process Engineering

(accepted).


108

Conference Contributions

Vorträge/Poster

[1] Abdel-Rahman, E. A. (1997) Evaluation of Mungbean seeds for technological

utilization. Workshop on Mungbean crop in Egypt. Oral presentation, (National

Center for Research, Cairo, Egypt, May 1997)


A. (2000) Effect of processing on the amino acid composition and protein quality of

mung bean seeds. First Mansora Conference of Food Science and Dairy Technology,

Oral presentation (Mansoura, Egypt, October. 2000).


A. (2000) Effect of some technological processes on the gross chemical composition

and mineral contents of mung bean seeds. First Mansoura Conference of Food

Science and Dairy Technology, Oral presentation (Mansoura, Egypt, October. 2000).

[4] Abdel-Rahman, E. A.; El-Syiad, S. and Kurz, T. (2006) Influence of high

dextran concentrations and molecular fractions on the rate of sucrose crystallization

in pure sucrose solutions. 13th World Congress of Food Science and Technology

‘Food is Life’, Poster presentation, (Nantes, France, September 2006).

[5] E. A. Abdel-Rahman; Q. Smejkal, R. Schick, El-Syiad, S. and Kurz, T. (2007)

Influence of dextran concentrations and molecular fractions on crystallization

kinetics in pure sucrose solutions. World Perspectives for Sugar Beet and Cane as a

Food and Energy Crop. Poster presentation (Sharm El Sheikh, Egypt March 2007).

[6] T. Kurz, E. A. Abdel-Rahman, R. Schick. (2007) Influence of dextran on

sucrose crystallization. CITS 23rd General Assembly, Oral presentation

(Warnemünde, Germany May 2007)

[7] E. A. Abdel-Rahman, R. Schick, T. Kurz (2007) Einfluss von Dextran auf die

Zuckerkristallisation. Frühsommertagung Technik, Pfeifer & Langen, Oral

presentation (Bielefeld, Germany 31th May to 01 June 2007).

109

Eidesstattliche Erklärung

Hiermit erkläre ich, dass ich die vorliegende Arbeit selbstständig und ohne fremde

Hilfe verfasst und andere als die angegebenen Hilfsmittel nicht verwendet habe.

Berlin, den 17. Juli 2007

El-Sayed Soliman

Investigations on the influence of dextran during beet ... · 4.6 Elucidation of crystallization...

Documents

Transcript of Investigations on the influence of dextran during beet ... · 4.6 Elucidation of crystallization...