Heat Transfer Rotating Surfaces

Copyright is owned by the Author of the thesis.
Permission is given for

a copy to be downloaded by an individual for the purpose of research and
private study only. The thesis may not be reproduced elsewhere without
the permission of the Author.
DEPARTMENT OF FOOD TECHNOLOGY
MASSEY UNIVERSITY
PALMERSTON NORTH
NEW ZEALAND
HEAT TRANSFER AND FOULING

IN FILM EVAPORATORS
WITH ROTATING SURFACES
A THESIS PRESENTED IN PARTIAL FULFILMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY IN FOOD TECHNOLOGY
AT MASSEY UNIVERSITY
HONG CHEN
B.E., M.Tech. (Honours)
1997
-------
Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces ii
ABSTRACT
A study was made on the heat transfer and fouling in thin film evaporators with
rotating surfaces. Both theoretical and experimental studies were carried out,
in order to gain a better understanding of these evaporators and their design
principles, so that this type of evaporator could be effectively used in an on-farm
milk evaporation system.
By using Nusselt-type assumptions, a theoretical model, which was used to

predict the liquid film thickness and heat transfer coefficients on the rotating
cone, was developed. The theoretical equations obtained revealed basic
relationships between the variables and provided a fundamental knowledge of
the liquid flow and heat transfer in the film evaporators with rotating surfaces.
The experimental studies on heat transfer were conducted on a Centritherm

evaporator, which is available commercially (400 half cone angle), a specially
made cone evaporator (100 half cone angle) and a falling film evaporator with
a rotating tube. Variables evaluated were the rotating speed, the cone angle, the
feed flow rate, the evaporating temperature, the temperature difference between
the steam condensing and the liquid evaporating temperatures, and sugar
concentration when sugar solution was used. The experimentally measured
overall heat transfer coefficients were compared with the theoretical values.
It was found that the measured overall heat transfer coefficients increased with
increase of the cone rotating speed, and with the rise of the liquid evaporating
temperature. The feed flow rate was found to have a more Significant effect on
the measured overall heat transfer coefficients in the falling film evaporator with
a rotating tube than that in the Centritherm and the cone evaporators. The
overall heat transfer coefficients decreased with increase of the concentration of
sugar solutions, mainly due to the increase of liquid viscosity. It was also found
Abstract III
that the measured overall heat transfer coefficients in the Centritherm evaporator
increased with an increase in temperature difference up to 30K (for water, 10% .
sugar solution and skim milk) and then decreased (for water and 10% sugar
solution). The formation of bubbles on the evaporating surface at high
temperature differences was likely to be cause of this effect.
Increase of the cone angle resulted in thinner liquid films and higher heat
transfer coefficients. This was reflected in the following experimental results: the
measured overall heat transfer coefficients in the falling film evaporator with a
rotating tube were slightly lower than those measured in the cone evaporator,
but much lower than those obtained in the Centritherm evaporator.
The experimental results showed that rotating the tube of a falling film
evaporator increased the overall heat transfer coefficient but the increase
obtained was very dependent on feed flow rate, and wC!.s not sufficient to justify
the use of this evaporator in the industry.
With the Centritherm evaporator, good agreement between theoretical and

experimental overall heat transfer coefficients as a function of the cone rotating
speed was obtained by using water. The theoretical model, however, does not
adequately describe the whole evaporation process at conditions other than those
assumed in the model, which are: laminar liquid film flow, and heat transfer by
conduction through the liquid film. It is suggested that waves existing in the
liquid film at high Reynolds numbers, and bubble formation on the heating
surface at high temperature difference, are the major reasons for the discrepancy
between theoretical and experimental results.
For the fouling study, the Centritherrn evaporator was mainly employed, and
three liquid systems: reconstituted skim milk, reconstituted whey solutions and
sweet cheese whey solution, were selected. It was found that no fouling was
detected after 6 hours' operation in the Centritherm evaporator when
Abstract iv
reconstituted skim milk and reconstituted whey solutions were used. This
indicates that the aggregated whey proteins, which are formed in the
manufacture of skim milk powder and whey powder, are less active in inducing
fouling. For this reason, only the sweet cheese whey solution was used in
further studies.
It was confirmed that fouling is strongly linked with the liquid evaporating
temperature and the temperature difference between steam condensing and
liquid evaporating temperatures. In general, the higher the evaporating
temperature and the temperature difference, the faster the deposition rate and
the greater the fouling on the surface. It was found that 72% Bovine Serum
Albumins (BSA) denatured after running the evaporator, at a evaporating
temperature of 70°C and a temperature difference of 20K, for 6 hours. Though
the content of BSA in whey solution is small, the denatured BSA could be easily
attached to the surface. By association with other depositable materials existing
in the whey solution, the thin layer of deposit could reduce the heat transfer
coefficients significantly. This was attributed to the lower thermal conductivity
of the deposited layer. Fouling was also found to be a function of the liquid
velocity. This effect was more significant at lower evaporation temperatures.
Increasing the rotating velocity would delay the formation of an initial layer and
reduce the rate of fouling.
It was also found that there was an induction period in the fouling curves when
the evaporating temperature was 60°C. The induction period was reduced when
new whey solutions were introduced into the evaporator. It proved the fact that
depositable materials are much more easily adsorbed on fouled or unclean
surfaces than on clean surfaces. The increase of fouling rate when new whey
solutions were introduced suggested that the concentration of activated
molecules in the solutions strongly affected the fouling process. A possible
mechanism of whey fouling on the rotating surface was proposed.

Abstract v
During this study, an attempt was made to develop a new type of evaporator
in which a vapour compressor would be integrated with the rotating surface.
This was unsuccessful due to the failure of compressing the vapour. Concerning
the on-farm evaporation system, which requires an evaporator with high
efficiency, compact, and minimum heat load to milk, it is suggested that a
rotating surface evaporator with the top cone angle close to 90° (like a disk
evaporator) would be optimum and worth to explore.

Heat Transfer and Fouling in Film Evaporators with Rotating surfaces vi
ACKNOWLEDGEMENTS
I wish to express my deepest appreciation and gratitude to my supervisor, Mr.

Selwyn Jebson, for his supervision, guidance, encouragement, patience and
friendship, and to my co-supervisor, Dr. Osvaldo Campanella, for his
suggestions, valuable advice and encouragement throughout this study at
Massey University.
I would like to extend my appreciation to Professor Peter Munro and Mr. Rod
Bennett, for their help and care, to Dr. Tuoc Trinh for his stimulating
discussions, to Mr. Byron Mckillop for his valuable assistance in setting up the
experimental apparatus, to Mr. Palatasa Havea for his assistance in testing whey
protein samples, to Mr. Mike Conlon, Mr. Alistar Young, Mr. Garry R adford,
Ms. June Latham, Mr. Steve Glassgow, Mr. Hank van Til, Mr. Mark Dorsey for
their help in providing technical assistance during the experimental work, to Dr.
P.K. Samal and his colleagues at Longburn Cheese Factory of Tui Milk Products
Company in Palmerston North for their efforts in arranging the whey solutions.
The staff of the Department of Food Technology were very helpful throughout
the duration of my study.
I am also grateful to Professor Alan Williams and Mrs Beverley Williams, for
their generous help, continuous care and friendship.
I would also like to mention the enjoyable company and moments provided by
my friends, fellows at Massey University, and the members of New Zealand
China Friendship Society, who in many ways contributed to the completion of
this work. Especially, I have to mention that when I was involved in an
unfortunate car accident, many friends provided their help for my family to go
Acknowledgements vii
through a difficult time. Their invaluable assistance is very much appreciated
and will be always remembered.
Finally, I would like to express my special thanks to my wife, Xinjun, for her
love, encouragement, patience and typing. During this period, my son, John,
and my daughter, Helen, came to this world and brought us a newly happy life.
I also thank my father, mother, sister, grandmother, parents-in-law as well as the
whole family for their understanding and constant support and helpfulness.
Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces Vlll
TABLE OF CONTENTS
ABSTRACT .... ............ ...... ........ ...... . . .......... ii

.
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . vi
TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
LIST OF FIGURES ................................ .......... xv

.
LIST OF TABLES .. ................ .............. .. .... ..... xix

.
NOMENCLATURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xx
LIST OF PUBLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxiv
PART I GENERAL INTRODUCTION
CHAPTER 1 INTRODUCTION . ... .................... .......... 1
1.1 Evaporation . . ............. ................ .......... 1
1.2 Evaporation technology used in the dairy industry . .......... 7
1.3 The thin film evaporator with rotating heating surface . . . . . . . . 8
1.4 Fouling in evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

.
1.5 On-farm evaporation system . . . . . . . . . . . . . . . . . . . . . . . . . .. 10
CHAPTER 2 OBJECTIVES . ........................... ......... 1 3
PART II HEAT TRANSFER IN EVAPORATORS WITH
ROTATING SURFACES
CHAPTER 3 HEAT TRANSFER IN THIN FILM EVAPORATORS -
LITERATURE REVIEW . .... .............. ........ .. 15
3.1 Liquid film flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.1.1 The flow regimes of the liquid film . . . . . . . . . . . . . . . 16
3.1.2 Flow features of the falling liquid films . . . . . . . . . . . . 18
3.2 Liquid boiling phenomena . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.1 Pool boiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.2.Flow boiling . . . . . . . . . . . . . . . . . . . . ' ... .......... 2 2
3.3 Heat transfer through a thin liquid film . . . . . . .. . . . . . . . .. . 24

3.4 Effect of the adjacent gas stream on the liquid film flow
and heat transfer. . ....................... ......... 27

- --- ---
Table of Contents ix
3.5 Heat transfer enhancement by rotating the surface .......... 29
3.6 Steam Condensation ................................. 3 2
3.6.1 Film C ondensation ............................ 3 3
3.6.2 Dropwise condensation ....... .......... ........ 3 4
3.6.3 Condensation in the presence of non-condensable
gases ..... . ........... .................

. . . 3 6
3.7 The development of evaporation technology in the
dairy industry .................................... 3 8
3.8 Thin film evaporators with rotating surfaces .. ............ . 43
3.8.1 The development of this type of evaporator .... .... . 43
3.8.2 The features of the rotating surface evaporator ..... .. 4 5
3.8.3 Industrial use of rotating surface evaporators ...... .. 48
CHAPTER 4 EXPERIMENTAL METHODS FOR DETERMINING
HEAT TRANSFER COEFFICIENT .... ............ .. . . 50
4.1 Objectives _ . .. .......................... ............ 50
4.2 Materials .... . ......................... .... ...... .. 50
4.. 2.1 Water ... . ...............

. . . . ..... . . ..... . . 50
4.2.2 Sugar s olution ...... ............. .... ....... .. 50
4.2.3 Reconstituted skim milk ........ ..............

. 50
4.2.4 Chemicals used for Cleaning-in-place (CIP) ......... 5 1
4.3 Equipment .. . .................................... .. 5 1
4.3.1 Centritherm evaporator. .. .......... .......... .. 5 1
4.3.2 Single-tube falling film evaporator with
rotating surface ......... ...... . .............. 5 4
4.3.3 Cone evaporator ................. .. ........... 5 6
4.4 The evaporation process in thin film evaporators .......... , 57
4.5 Instrumentation ........... .... .......... .......... .. 59
4.5.1 Steam regulator .............................. 59
4.5.2 Flowmeter ........................ . . ........ 59
4.5.3 Temperature measurement ...................... 60

Table of Contents· x
4.5.4 Pressure measurement ....... .................

. 60
4.5.5 Rotating speed measurement ..... .. .... ......

. . . 60
4.6 The variables and their ranges ......................... 6 1
4.7 Operating procedure of thin film evaporator equipment . ..... 6 2
4.7.1 Starting ............. ..... .. ................. 6 2
4.7.2 Cleaning .................................... 6 2
4.7.3 Stopping ..
. ................................ . 6 3
4.7.4 Operating cautions ............. ..............

. 6 3
4.8 Experimental Procedure. ...... .. . ... .................. 6 4
4.9 Data processing . ................. ... . ............... 6 5
CHAPTER 5 DEVELOPMENT OF A THEORETICAL MODEL
TO DETERMINE HEAT TRANSFER COEFFICIENTS
ON R OTATING CONE SURFACES ................... 67
5.1 Introduction. ....................................... 67
5.2 Theory .. . .. .................... ............ . . .. ... 67
5.2.1 Liquid film flow and energy analysis ......... .... .

67
5.2.2 Determination of the local heat transfer coefficient . ... 7 4
5.2.3 Effect of liquid evaporation . . ........... .... ..... 7 5
5.2.4 Overall heat transfer coefficient calculation. ...... ... 7 5
5.2.5 The relationship between rotational Nusselt
number and Reynolds number . .......... .. ..... 77
5.3 Calculation procedure ..................

. .... .... .. ... 7 8
5.4 The determination of the physical properties of the liquids used 78
CHAPTER 6 HEAT TRANSFER IN THIN FILM EVAPORATORS-
RESULTS AND DISCUSSION . .. ................... .. 80
6.1 Theoretical model results ........................... ... 80
6.2 Experimental results and their comparison with
those theoretically calculated with the model ............ 8 8
6.2.1 Centritherm evaporator. .. .... .... ...... ... . .... 90
6.2.1.1 Temperature difference ... . . ............. 9 1
6.2.1.2 Rotating speed . .............. ........ 9 3 .

Table of Contents xi
6.2.1.3 Feed flow . . . . . . . . . . . . . .

. . . . . . . . . . . . . . 95
6.2.1.4 Evaporating temperature . . . . . . . . . . . . . . . . 100
.
6 .2.1.5 Experiments of sugar solutions at

different concentrations . . . . . . . . . . . . . . . . . . 103
6.2.2 Cone evaporator . . . . . . . . . . . .. . . . . .. . . . . .. . . . . . 106
6.2.2.1 Temperature difference . . . . . . . . . .
. . . . . . . 106 .
6 .2.2.2 Rotating speed . ... .. ..... ... . ... .. . .... 106

6 .2 .2 .3 Feed flow . . . ... . . .. . . . . . . . . . . . . . . . . . . . 109
6.2.2.4 Evaporating temperature . . . . . . . . . . . . . . . . . 109
6 .2.3 Falling film evaporator with a rotating tube . . . . . . . . . 109
6.2 .3.1 Reynolds number . . . . . .. . . . . . . . . . . . . . . . . 112 .
6.2.3.2 Rotating speed . . . . . . . . . . . . . . .. . . . . . . . . . 114

6.2.3.3 Evaporating temperature . . . . . .. . . . . . . . . . 116
6 .2.3.4 Feed temperature . . . . . . ... . . . . . . . . . . . . . 116
6 .2 .3.5 Temperature difference . . . . . . . . . . . . . . . . . . 119
6.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1
PART III FOULING IN EVAPORATORS WITH ROTATING SURFACES
CHAPTER 7 LITERATURE REVIEWS OF FOULING IN
NITLKPROCESSING ............................. . . 12 3
7.1 The costs of fouling .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . 12 3
7.2 The types of fouling ..
. . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. 12 5
7.3 The research history of fouling in dairy industry . _ . . . . . . . . . . . 12 6
7.4 Composition and structure of milk fouling deposit . . . . . . . . . . 12 8
7 .5 Factors affecting milk fouling on the heating surface . . . . . . . . . 131
7 .5.1 Milk properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
7.5.1.1 Compositional variations with season . . . . . . . 131
7.5 .1.2 pH values of milk . . . . . . . . . . . . . . . . . . . . . 132 .
7.5.1.3 The amino nitrogen level of the milk . . . . . . . . 132

7.5 .1.4 Calcium phosphate . . . . . . . . . . . . . . . . . . . . . 132
7.5 .1.5 Air content in milk . . . . . . . . . . . . . . . . . . . . . 133
7.5.1.6 Concentration of milk . . . . . . . . . .. . . . . . . . . 135
Table of Contents xii
7.5.2 Plant construction and operation ......... ..... . . . 135

7. 5.2.1 Storage time .... ....... ......... ... . 135
. . .
7.5.2 .2 Preheating . . .. ...................... 13 6

. .
7 .5.2.3 Absolute liquid velocity .. ........ . ... .... 13 6

7.5.2 .4 Surface conditions .... .... . ....... .. . 137 . . .
7 .5.2.5 Temperature . .... .... . . . .. . .. .. . . . . .. 138

.
7.5.2 . 6 Processing time .. ..... . . ..... . .... . 138 . . . .
7. 6 Possible mechanisms for milk fouling on the heating surface .. 140

7.7 The initial deposited layer of milk fouling . .. .......... . . 143 . .
7.8 The rate-controlling processes in milk fouling . ......... .... 145

7.9 Whey proteins and their denaturation ..... . .............. 14 6
7.9.1 �-lactoglobulin (�-lg) ... ... .... ......... ... ... 147
. .
7.9.2 a-lactalbumin (a-Ia) .......... ... ... .... . ... . 147 . .
7.9.3 Bovine serum albumin (BSA) ..... ... . . . ......... 148

7 .9.4 Immunoglobulins (Ig) ... .. . . ...... ..... ... ... 148. .
7 .9.5 Whey protein structure ...... . . . . ............ . .. 148

7 .9 . 6 Whey protein denaturation and aggregation .... .. . . 149 .
7.10 The relationship of whey protein denaturation and

milk fouling .... . ..... .. ... .... ... .......
. . . . .... 149
7.11 Conclusions . ...... . ..... . . . ... ... .............. 150
. . .
CHAPTER 8 FOULING IN EVAPORATORS - MATERIALS AND
METHODS . .............................. . .... 152 . .
8.1 Introduction...... . .. ..... . ..... .. . ............... 152 . .
8.2 Materials .. .. ........ ...... ..... .... .......... ... 152

.
8.2.1 Equipment ... ... ......... ... . .......

. . . ..... 152
8.2.2 Model Solutions .... ............. ........... 152
. .
8.2.3 Chemicals used for cleaning ... . .... . ............ 153

8.3 Methods ............ ... ............. . ............. 154
8.3.1 The Centritherm evaporator and the falling film
evaporator . . .. ..... .. ..... ... ..... .. .. ..... 154
8.3.2 The recycle system . ..... . ... .. . . . . .. ... ... 154

. . . .
Table of Contents xiii
8.3.3 The examination of denatured whey proteins ........ 155

8.4 Instrumentation ..... ...........
. . . . . . .... ... .... . 155
. . .
8.5 Experimental ...... ..... ............

. . . . .... ... . . . . . 155
8.5.1 Experimental variables ......................... 15 5
8.5.2 Experiment design ........ . .... ............ 15 6 . . .
8.5.3 Experimental Procedure ........... .... ....... 15 7 . .
8.5.4 Evaporator cleaning ... .............. . . . ....... 15 7

8. 6 Theory and data processing ........................... 15 8
CHAPTER 9 FOULING IN EVAPORATOR--RESULTS AND
DISCUSSIONS ................... ............... 1 60 .
9.1 Preliminary experimental results .. ...... .. . . . . . ......... 1 60

9.2 Effects of evaporation temperature and temperature
difference on fouling............... .. .. ........ .... 1 62
9.3 Visual observation of the whey deposits on the surface ....... 1 67

9.4 Effect of the rotating speed on fouling................ ... 1 67 .
9.5 The interaction of variables on the fouling resistance ......... 17 0

9. 6 The building up of the deposits ...... ........... ...... 1 7 2 . .
9.7 The possible fouling mechanism of whey solutions o n the

rotating surface............ ................. . .. 1 7 3
. . .
9.8 The evaporation mechanism on a fouled heat transfer surface .. 1 7 3

9.9 Conclusions ....... .... ........ . ....
. . . . . . . ........ 1 7 5 .
CHAPTER 10 OVERALL DISCUSSION AND
RECOMMENDATIONS.................. ..... . . .. 17 7
10.1 Overall discussion ............ .. ... ... ........ 17 7 . . . . .
10.1.1 Background of the PhD programme ....... . . . . . . 17 7

.
10.1.2 Liquid film flow and heat transfer on the

rotating surface................ ............. 1 7 8 .
10.1.3 Heat transfer in the Centritherm and

the cone evaporators . ...... ... ............. 178
. . .
10.1.4 Heat transfer in th e falling film evaporator

with rotating tube... .. ..... . . . . . . . . .. ... . . . . . 181
Table of Contents xiv
10 .1.5 Fouling with whey solution in the

Centritherm evaporator . . . . . . . . . . . . . . . .. . . . . . . 181
10.2 Recommendations for the further work . . . . . . . . . . . . . . . . . . 182
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
APPENDICES
Appendix I Experimental results of heat transfer in
evaporators . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 14
Appendix n Calculation of surface areas on the cone . . . . . . . . 225
Appendix ill Correlations of physical properties for
water, sugar solution and skim milk . . . . . . . . 226
Appendix N Results of numerical calculation . . . . . . . .. ... . . . 2 30
Appendix V An example of determining the uncertainty
for overall heat transfer coefficient . .. . . . . . . . 2 43

Appendix VI Polyacrylamide Gel Electrophoresis (PAGE) . . . . . 2 44
.
Appendix vn Experimental results of fouling in

Centritherm evaporator . . . . . . . . . . . . . . . . . . 2 45
Appendix VITI A typical native-PAGE patterns of whey protein
solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces xv
LIST OF FIGURES
Figure 1.1 The different types of evaporator ..................... 3 & 4

Figure 1.2 The proposed on-farm evaporation system ................ 12
Figure 3.1 The boiling curve for water at atmospheric pressure......... 2 1
Figure 3.2 The different designs of the film evaporator with
rotating surfaces ............. ............ .. ....... 46
Figure 4.1 The whole apparatus of the Centritherm evaporator and
the falling film evaporator with a rotating tube .......... 5 2
Figure 4.2 Schematic diagram of the Centritherm evaporator system 53
Figure 4.3 Schematic of the single-tube falling film evaporator
with a rotating surface ......... ................... 5 5
.
Figure 4.4 Schematic diagram of the feed tube ............... ...... 5 6 .
Figure 4.5 Schematic diagram of cone evaporator and the

dimensions of the cone ... ................. .... ..... 5 8
Figure 5.1 Schematic diagram of the system used in the theoretical analysis 6 8
Figure 5.2 Flow diagram for the numerical calculation of the
overall heat transfer coefficient . . . . . . . . . . . . . . . . . . . . . . . 7 9
Figure 6.1 Dimensionless film thickness on a rotating cone at
different cone angles....... .................. ...... 81
Figure 6.2 Dimensionless film thickness on a rotating cone at
different feed flows.. ....................... .. .. ... 82
Figure 6.3 Dimensionless film thickness versus rotation parameter
at different cone angles.......... ......... ........ , 84 .
Figure 6.4 Effect of rotating speed on the Nusselt number at

different cone angles...................... .. .. ... .. 85
Figure 6.5 Calculated heat transfer coefficients on the rotating cone in
the Centritherm evaporator ................ .. .. ..... 86
.
Figure 6.6 Calculated local surface temperatures, liquid film

thickness and Reynolds number along the cone . . ........ 87
List of Figures xvi
Figure 6 .7 Recorded temperatures during an experimental

run in the Centritherm evaporator . . . . . . . . . . .
. . . . . . . . . 89
Figure 6.8 Effect of temperature difference on the overall heat transfer
coefficient for water, 20% sugar solution and skim milk
in the Centritherm evaporator . . . . . . . . . . . . ...
. . . . . . . . 92
Figure 6 .9 Effect of the cone rotating speed on the overall heat transfer
coefficient for water, 20% sugar solution and skim milk in the
Centritherm evaporator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Figure 6 .10 Effect of the feed flow rate on the overall heat transfer
coefficient for water, 20% sugar solution and skim
milk in the Centritherm evaporator . . . . .. . . . . . . . . . . . . . 96
Figure 6 .11 Effect of the feed flow rate and the rotational speeds
on the overall heat transfer coefficient for water
in the Centritherm evaporator . . . . . . . . .. . . . .... . . . . . . 98
Figure 6 .12 Effect o f the feed flow rate on the overall heat transfer
coefficient for water in the Centritherm evaporator . . . . . . . 99
Figure 6 .13 Effect of evaporating temperatures on the overall heat
transfer coefficient in the Centritherm evaporator . . . . . . . . . 101
Figure 6.14 Effect of the temperatures difference on the overall
heat transfer coefficient for sugar solutions in
the Centritherm evaporator . . . . . . . . . .. . . . . . . . . . . . . . . . 104
Figure 6 .15 Effect of sugar concentration on the overall heat
transfer coefficient in the Centritherm evaporator . . . . . . . . . 105
Figure 6 .16 Effect of the temperatures difference on the overall heat
transfer coefficient for water in a cone evaporator . . . . . . . 10 7 .
Figure 6.17 Effect of rotating speed on the overall heat transfer

coefficient at different evaporating temperature
in a cone evaporator with water . . . . . . . . . . . .... . . . . . . 108 .
Figure 6 .18 Effect of flow rates on the overall heat transfer

coefficient for water in a cone evaporator . . . . . . . . . . . . . . . 110
List of Figures xvii
Figure 6 .19 Effect of evaporating temperature on the overall

heat transfer coefficient for water in a
cone evaporator .... .. . . . . . . . .. . . . .... . . . . .. . . .. .. 111
Figure 6 .20 Effect of the Reynolds number on the overall heat
transfer coefficient of a rotating tube
falling film evaporator . . . .. . .. .. . . . . . .. .. . . . . ...... 113
Figure 6.21 Effect of the tube rotating speed on the overall heat
transfer coefficient in a rotating tube falling film
evaporator at different Reynolds numbers .. . .. ...... . ... 115
Figure 6.22 Effect of the evaporating temperature on overall heat
transfer coefficients in a rotating tube falling
film evaporator ...... ... . .. . . .. . ... ...... . . . ...... 117
Figure 6 .23 Effect of the feed temperature on the overall heat
transfer coefficient in a rotating tube falling
film evaporator . ... .. . . . .. . . . . . . . .. ... ... . . . ... ... 118
Figure 6 .24 Effect of the tube rotating speed on the overall heat
transfer coefficient in a rotating tube falling film
evaporator at different temperature differences .. . . ..... . 120
Figure 7.1 The schematic presentation of the milk fouling process
induced by an air bubble at a hot stainless steel surface . . .. 134
Figure 7.2 Idealised fouling curves .. . . .. . . . . . . . .. .. .. ... . ... .... 139
Figure 7.3 Schematic presentation of the fouling mechanisms during
heating of whey and milk . ....... .. .. .. ... . . . . . . . ... 142
Figure 9 .1 Fouling resistance as a function of time for recycled
sweet cheese whey in the Centritherm evaporator at
an evaporating temperature of 6 0°C and a rotating
speed of 105 rad/s . .... . .. . . ..... . . ... .... . . .. ... . 16 3
Figure 9 .2 Fouling resistance as an function of time for recycled
sweet cheese whey in the Centritherm evaporator at
an evaporating temperature of 7 0°C and a rotating
speed of 105 rad/s .. . .. . .. ...... .. .... .. .. .. .. .... 16 4
List of Figures xviii
Figure 9.3 Fouling resistance as an function of time at different

evaporating temperatures for recycled sweet cheese
whey in the Centritherm evaporator . .. . . .. . . . .. . .. . .. . 1 65
Figure 9.4 The fouled rotating surface of the Centritherm evaporator . .. 1 68
Figure 9.5 Fouling resistance as an function of time for recycled sweet
cheese whey in the Centritherm evaporator at an
evaporating temperature of 7 0°C and different
rotating speeds. . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . . . . . 1 6 9
Figure 9. 6 Comparison of fouling resistances in the falling film
evaporator and the Centritherm evaporator for recycled
sweet cheese whey at a temperature difference of 20 °C . . . .. 1 7 1
Figure 9.7 Fouling resistance as a function of time for recycled sweet
cheese whey, when a new whey solution is introduced . .. . 1 7 4
Figure 9.8 Overall heat transfer coefficient as a function of
temperature difference for different liquids
and clean and fouled rotating surfaces in
the Centritherm evaporator . . . . . . . . . . . . . . . . . .. . . . . . . . 1 7 6
Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces xix
LIST OF TABLES
Table 4.1 Ranges of experimental variables ......... . . . .. . . . ...... 61 .
Table 6.1 Physical properties of water as a function of temperature . . . .. 10 2

Table 6. 2 Physical properties of a 20% sugar solution as a
function of temperature .. ... . . .. .... .. . . . ... ... . . . . 10 2
Table 6.3 Physical properties of skim milk as a function of temperature . 10 2 .
Table 7.1 The composition of milk deposits in a pasteurizer and a

sterilizer . . . .. . . . .
. . . . . ......... . ... . . . . .. .. 12 9
. . . . .
Table 7.2 The protein composition of milk deposits in a pasteurizer

and a sterilizer ...... . .. ..... . ..
. . . . ... . .... .. .. . 12 9
.
Table 8 .1 Ranges of experimental variables .. .. . . ... .. ... .. .. ... . . 15 5

.
Table 8 .2 The arranged trials with selected variables .. . ... . .. .. .. . .. . 15 6

Table .
9 1 Summary of preliminary experimental results after 6 hours
evaporationnnuting . . . . . ... . .. . . . . . . . . . . . .
. . . . . . . . 161
Table 9.2 Percentage loss of native whey protein during evaporation
at a temperature of 7 0 °C and a temperature difference
of 20 K . . . . . . . . .. ... . . . .. . . . . . . . . . . . . . . . . .. . ..... 166
Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces xx
NOMENCLATURE
Roman letters
A heat transfer area (m2)
Al liquid side heat transfer area (m2)
Am average heat transfer area (m2), defined as: (As-AI)/ln(AJ AI)
As steam side heat transfer area (m2)
a acceleration due to rotating (m/S2)
C constant
C specific heat (kJ/kg.K)
p
d inside tube diameter (m)
do outside diameter on the top of cone (m)
D outside tube diameter (m)
Do outside diameter at the bottom of cone (m)

Fr Froude number, defined as: [u2/(g.d)]
g acceleration due to gravity (m/ S2)
g' corrected gravitational acceleration, defined as g.cosB (m/S2)
hfg latent heat of vapour condensation (kJ/kg)
'
h fg effective latent heat of vapour condensation (kJ/kg), defined as:
hfg[1+0 .68Cp(Ts - Tws)/hfg]
hs steam side heat transfer coefficient (kW/m2.K)
hI liquid side heat transfer coefficient (kW/m2. K)
'
hi corrected liquid side heat transfer coefficient (kW/m2. K)
Ho initial overall heat transfer coefficient (kW/m2.K)
Heal calculated overall heat transfer coefficient (kW/m2. K)
Hex measured overall heat transfer coefficient (kW/m2.K)
p
H(t) measured overall heat transfer coefficient at time t (kW/m2.K)
Ja Jacob number, defined as: C (Ts-Tw)/h'fg
p
k thermal conductivity of liquid (W/m.K)
kd thermal conductivity of fouling layer (W/m.K)
kss thermal conductivity of sugar solution (W/m.K)
Nomenclature
xxi
kw thermal conductivity of wall (W/m.K)

L characteristic distance (m)
L1 distance from the vertex to the apex of a truncated cone (m)
I length of the cone (mm)
Nu Nusselt number, defined as: hlL/k
Nu' modified Nusselt number, defined as eq (3-4)
NtIro rotational Nusselt number, defined as eq (5-37)
Pr Prandtl number, defined as: c;,p/k
Q amount of heat (kW)
Qf feed flow (m3/s)

q heat flow (kW/m2)
r inner radius on the any position of cone (m)
rj inner radius on the top of cone (m)
Rt fouling resistance (m2.K/kW)
Rt(t) fouling resistance at time t (m2.K/kW)
Rea axial flow Reynolds number, defined as: Du/v
Rer rotational flow Reynolds number, defined as: D20/v
R; inner radius at the bottom of cone (m)
Ro rotation parameter, defined by eq (5-19)
Urn characteristic velocity (m/s)
Ux component of liquid velocity in X direction (m/s)
Uy component of liquid velocity in Y direction (m/s)
Uq> component of liquid velocity in cp direction (m/s)
u liquid flow velocity (m/s)
T temperature (oC)
Tevp evaporating temperature (oC)
TI liquid temperature(oC)
Ts steam temperature (oC)
Tsat saturation temperature (K)
Tw wall temperature (oC)

Twi surface temperature of cone on liquid side (oC)
Nomenclature xxii
Tws surface temperature of cone on steam side (OC)

AT temperature difference between the evaporating and the steam
condensing temperatures (K)
temperature difference between the wall and liquid saturation
temperature (K)
Weal calculated flow rate of vapour condensate (kg/s)
measured flow rate of vapour condensate (kg/s)
Weber number, defined as: [(p.u2. 8 /0')112]
x downward distance along cone surface (m)
y distance from surface (m)
Greek symbols
a thermal diffusitivity (m2/s)

half angle of the cone
thickness of liquid film (mm)
thickness of the fouling layer (mm)
thickness of wall (mm)
dynamic viscosity (Pa.s)
v kinematic viscosity (m2/s)
p density of liquid (kg/m3)
Pv density of vapour (kg/m3)
Pss density of sugar solution (kg/m3)

n angular velocity (rad/s)
<p angular co-ordinate

r local mass flow rate in the film per unit width of surface (kg/m.s)
'ty radial shear stress (N/m2)
x
tangential shear stress (N/m2)

surface tension (N/m)
Superscripts
*
dimensionless quantity
Nomenclature xxiii
Abbreviations
BPE Boiling Point Elevation
CIP Cleaning In Place
DM Dry Matter
ET Evaporation Temperature
FF Feed Flow
FFE Falling Film Evaporator
FT Feed Temperature
PAGE Polyacrylamide Gel Electrophoresis
RS Rotating Speed
TS Total Solids
Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces xxiv
LIST OF PUBLICATIONS
H.Chen, RS.Jebson and D.H.Campanella, 19 9 3, Factor Affecting Heat Transfer

in the Centritherm Evaporator, Proceedings, the APCChE/CHEMECA 1993
,
Conference, VoL3, pp 227 -233, Melbourne, Australia.
H.Chen, RS.Jebson & O.H.Campanella, 19 9 4, Heat Transfer Coefficients for

Evaporation from the Inner Surface of the Rotating Cone, Proceedings, CHEMECA
1994 Conference, Vo1.2, pp 69 5-7 02, Perth, Australia.
H.Chen, RS.Jebson & O.H.Campanella, 19 9 4, Performance of the Centritherm

Evaporator for Concentration of High Viscosity Solutions, Proceedings of the
Inaugural New Zealand Postgraduate Conference for Engineering and Technology
Students, Department of Production Technology, Massey University, pp 60- 65.
RS.Jebson, H.Chen, & D.H.Campanella, 19 9 5, Heat Transfer Coefficients on the

Inner Surface of the Rotating Cone, Proceedings, 1995 Asian Pacific Confederation
of Chemical Engineer, Vo1.2, pp 125 -13 6, Taibei Taiwan.
H.Chen, R.S.Jebson & O.H.Campanella, 19 9 6, Fouling in the Centritherm

Evaporator with Whey Solutions, Proceeding of IPENZ 1996 Conference, Vo1.2, Part
2, pp 308-313, Dunedin, New Zealand.
RS.Jebson, H.Chen & O.H.Campanella, 19 9 7 , Study on a Rotating Tube Falling

Film Evaporator, Proceeding of IPENZ 1997 Conference, Vo1.2, pp 41-45,
Wellington, New Zealand.
H.Chen, RS.Jebson & O.H.Campanella, 19 9 7 , Determination of Heat Transfer

Coefficients in Rotating Cone Evaporators Part I, Food and Bioproducts Processing,
Trans IChemE, Vol. 75, Part C, pp 17 -22.
List of Publications xxv
RS. Jebson and H.Chen, 1 9 97 , Performances of Falling Film Evaporators on

Whole Milk and a Comparison with Performance on Shim Milk, Journal of Dairy
Research, 64, pp 5 7- 67 .
H.Chen, RS.Jebson & O.H.Campanella, 1 9 9 7, Determination of Heat Transfer

coefficients in Rotating Cone Evaporators Part II, a paper to be submitted to Food
and Bioproducts Processing, Trans IChemE, Part C.
H.Chen, RS. Jebson & O.H.Campanella, 1 9 9 7, Fouling of Heat Transfer Surface

by Whey Solution in Rotating Film Evaporators, a paper to be submitted to
International Journal of Food Science and Engineering.
PART I
GENERAL INTRODUCTION
Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces 1
CHAPTER 1
INTRODUCTION
1.1 Evaporation
Evaporation is the process of conversion of a liquid into its vapour. It is also the name
given to a widely used unit operation in which solutions are concentrated by evaporating
a volatile solvent. It is one of the oldest means ever to have been adopted for
separating liquid mixtures by heat on an industrial scale (Billet, 1989) and is still an
important technique used for the removal of water and sometimes other liquids in
industry.
Evaporation often encroaches upon the other unit operations known as distillation,
drying and crystallisation. Evaporation is distinguished from distillation in that no
attempt is made to separate the components of the vapours in evaporation. Evaporation
differs from drying in that the residue is always a liquid, which means that the heat
must be transferred in the evaporator to a solution, although the desired final products
may be in a solid form. Evaporation is differentiated from crystallisation in that the
former is concerned with concentrating a solution rather than producing or building
crystals (Minton, 1986).
In an evaporator, heat must be supplied to the liquid through a heat exchange surface,
and the vapour and liquid separated. An evaporation process is therefore both a heat
exchange operation and a vapour-liquid separation. In industry, the heat is usually
supplied by condensation of steam. In the food industry, the evaporated portion is
usually pure water, although some other volatile components of the solutions are also
partially or totally vaporized.

Chapter 1 I ntroduction 2
Evaporation is extensively used in the food industry for the following reasons:
( 1 ) To reduce storage and shipping volumes of liquid foods, hence to reduce packaging,
storage, transport and distribution costs;
(2) To pre-concentrate foods prior to drying, freezing or sterilisation and hence to save
energy in subsequent operations;
(3) To increase the solids content of a food and hence to provide storage stability for
it by reducing water activity;
(4) To provide concentrates that are convenient to use for consumer and for further
processing.
Basically an evaporation system consists of:
(1) A heat exchanger to supply sensible heat to raise the liquid to its boiling point and
provide the latent heat of vaporisation;
(2) A separator in which the vapour is separated from the concentrated liquid phase;
(3) A condenser to condense the vapour and remove the condensate from the system;
(4) A vacuum device to withdraw non-condensable gases and maintain a constant
evaporating temperature.
There is not any single type of evaporator which could be satisfactory for all
applications and different kinds of feed. In general, the different designs for the heat
exchangers result in different types of the evaporators. Evaporators used for
concentration of liquids in the industry are normally classified as follows (Billet, 1 989):
(a) Natural circulation evaporator
(b) Forced circulation evaporator
(c) Climbing film evaporator
(d) Falling film evaporator
(e) Flash evaporator
(f) Horizontal heated tubes evaporator
(g) Thin film evaporator
Figure 1.1 shows schematic diagrams of these types of evaporators.

Chapter 1 Introduction 3
V
.....
-
S �
T
S
C
I
T
F C
-+--
p F p
(a) Natural circulation evaporator (b) Forced recirculation evaporator
(c) Climbing film evaporator (d) Falling film evaporator
F - Feed P - Product (concentrate)

V - Vapour C - Condensate
S - Steam
Figure 1.1 The different types of evaporator (continued on the next page)
F v
v t
F -+ - - - - - -
s
JC-
'-
- - - --
c
- - - - -
�
p
(e) Flash evaporator (f) Horizontal heated tubes evaporator
PF
v
tJ
F
s
(g-l) Wipers thin film evaporator (g-2) Rotating surface thin film evaporator
F - Feed P - Product (concentrate)

V - Vapour C - Condensate
S - Steam
Figure 1.1 The different types of evaporator (continued)

The natural circulation evaporator (Fig. 1 . 1 a) was the first type of evaporator to receive
wide acceptability for the industrial concentration of liquids. It became so common that
this type was sometimes known as the standard evaporator (Minton, 1986). The vertical
tube bundle with a central downtake is located inside a steam chest enclosed by a
cylindrical shell. The circulation of liquid (natural convection) past the heating surface
is induced by boiling, which improves the heat transfer.
The forced circulation evaporator (Fig. 1 . 1 b) was developed from the standard
evaporator. It has a vertical heated tube bundle located in an external heater separated
from the downtake, which made it possible to pump the liquid upwards through the tube
bundle. The fast liquid velocities improve the heat transfer and reduce the degree of
fouling. This evaporator can also be operated by natural circulation.
The climbing film evaporator (Fig. 1.1 c) has long heating tubes (up to 7 m) and the
feed is pumped into the bottom of the vertical tubes. The production of vapour bubbles
carries the liquid up the tubes, which results in an upwardly flowing film of liquid on
the top of tube wall, with a high velocity vapour core. This evaporator has relatively
high heat transfer coefficients and little or no re-circulation.
The falling film evaporator (Fig. 1 . 1 d) has vertical heating tubes and the feed is
carefully distributed to the top of the tubes. The feed forms a thin film and flows down
on the inner surfaces of the tubes. The vapour and the concentrate leaves at the bottom
of the tubes. This evaporator works successfully with wall temperatures of 2-3°C above
the feed boiling temperature and has a low residence time.
The flash evaporator (Fig. 1 . 1 e) is a low pressure chamber without heaters, in which
a hot feed is introduced and flash evaporation of the liquid occurs. This type of
evaporator is the most suitable for solutions which are prone to salting or scaling and
are very corrosive . It has been successfully used for producing potable water from
_
seawater (Al-Zubaidi, 1 987).

The horizontal heated tube evaporator (Fig. 1 . 1 t) has heating tubes (within which steam
condenses) immersed in the liquid to be evaporated. It is used in the production of
potable water.
The thin film evaporators have been designed to deal with viscous solutions and have
low residence times. There are two basic types: the one shown in Figure 1 . 1 g- l , in
which wipers are mounted to scrape the liquid off the heating surface, the other one
showl1. in Figure 1 . 1 g-2, in which a rotating heating surface is included. The later one
is further detailed in section 1 .3. The study carried out in this work was on the thin
film evaporator with rotating surface as that illustrated in Figure 1 . 1 g-2.
Although the underlying principle of evaporation is simple, the detailed liquid flow
patterns and the evaporation mechanisms on the heating surfaces are not very well
understood. The complications in the evaporation process also arise as a result of the
variety of products processed which have different properties. The behaviour of the
liquid during evaporation introduces a number of aspects of which consideration must
be given in designing and selecting evaporators. In other words, many parameters that
depend on the characteristic properties of the evaporating liquid govern the choice of
equipment and the operating conditions. They generally include: (a) the liquid's
viscosity in relation to the degree of concentration required; (b) the tendency to salting,
scaling and fouling; (c) the hold up time for thermally unstable products. These
properties are also affected by the physical properties of the processing liquids,
including thermal conductivity, density and thermal capacity. For dairy products, the
particular parameters to be considered are creaming, souring, foaming, undesirable
changes in viscosity and taste, discolouring, fouling and bum-on at the surface of the
heaters.
1.2 Evaporation technology used in the dairy industry
Evaporation technology is intensively used in production of milk powder, condensed
milk, lactose, whey protein concentrates. In the dairy industry, evaporators are often
used in conjunction with spray drying units. In these cases the concentration process
in the evaporator is very important because the specific energy consumption· of the
evaporator is much lower than that of the dryer. At the same time, by far the largest
part of moisture (up to 90%) is removed in the evaporator (GrllSnlund, 1 984, Fluck,
1 988). Evaporation is also claimed to be a compulsory step in powder processing
because milk powder produced from evaporated milk has longer shelf life and larger
particles with a smaller amount of occluded air (Carie, 1 994).
It has been calculated that in the manufacture of milk powder, 50% of the energy
consumed is required at the evaporation stage. In New Zealand 1 % of the total energy
from fossil fuels is consumed in evaporation processes (Jebson, 1 989). Therefore, the
study of the evaporation process in order to obtain a better understanding of the heat
transfer involved in it has an important industrial value.
Due to the heat-sensitive nature of milk and the energy costs rising, there are two
aspects of the evaporation process which have received much attention during its
development. One is to improve the quality of the finished product following an
evaporation process, to enable a better price to be obtained from it. Another is to
reduce the cost of removing water by this means.
Various types of evaporators have been used over the years in the dairy industry. The
natural circulation evaporators (Fig. 1 . 1 a and b) were the standard plant for many years.
These were superseded first by the climbing film evaporator (Fig. 1.1 c), and then by
the falling film evaporator (Fig. 1 . 1 d), which gave rise to a much improved product
quality due to considerably reduced residence times. Nowadays, the dominant
evaporators used in New Zealand dairy industry are the falling film evaporators.
Ch apter 1 Introduction 8
Although there are other concentration techniques (membrane technology and freeze
concentration etc.) being developed, evaporation is, and no doubt will remain, a major
technique used for the removal of water in the dairy industry (Fergusson, 1 989). In
New Zealand, evaporation is the only process used for the concentration of milk
(Mackereth, 1 995).
1.3 The thin film evaporator with rotating heating surface
Although evaporators are tending to become larger and despite the efforts to improve
the efficiency of the process, the evaporation process is still costly. It is therefore
believed that scope exists for the development of a compact, low cost, evaporator, linked
to a cost-effective heat transfer surface (Reay, 1 99 1 ).
Rotating the surface on which evaporation takes place appears to be a logical way of
developing this type of evaporator. In fact, this type of evaporator has been used for
decades. Thin-film evaporators with rotating heating surfaces provide an alternative of
small to medium capacities in the food, chemical, and pharmaceutical industries.
This kind of evaporator has an externally heated rotor in the form of a truncated cone.
The liquid to be evaporated is fed onto the inner surface of the cone, and the steam
condenses on the outer surface. Under the centrifugal field, the liquid film on the inner
surface moves very fast in a radial direction, consequently the liquid film becomes
extremely thin. Therefore, the heat transfer resistance of liquid film is small. On the
other hand, the condensate from the steam is impelled off from the outer surface of the
cone by the action of the centrifugal force, which results in more areas being exposed
for steam to be condensed. Therefore, the measured overall heat transfer coefficients
on the rotating cone surface could be as high as 1 0 kW/m2K (Chen et aI., 1 993, Anon,
1 990) compared to 2-3 kW/m2K for a falling film evaporator (Chen, 1 992).
This type of evaporator is especially suitable for heat sensitive biological materials. As
the liquid film passes the heating surface rapidly, the liquid residence time in the actual
evaporation zone may be only a fraction of a second (Billet 1 989). This short time is
sufficient to evaporate the liquid to required concentration and minimising deterioration
of the product' s characteristics, such as taste, colour, protein and vitamin content
(Mannheim and Paasy, 1 974).
This type of evaporator can be also used to concentrate viscous solutions, because the
liquid can be distributed and moved along the heating surface with the assistance of
centrifugal force generated in the rotating system. The viscosity of the final product
produced in this type of evaporators could be up to 20 Pa.s (Mannheim and Paasy,
1 974).
Applications of this type of evaporator have been found in many industries, but there
is a lack of basic studies dealing with the operation and performance of this type of
evaporator.
1.4 Fouling in evaporation
Fouling in food process plants is an example of an especially severe problem which
affects most heat transfer equipment. Deposit material on the heat transfer surface acts
as a barrier to heat transfer, decreases the overall heat transfer coefficient, increases the
overall plant pressure drop and poses a threat to plant sterility. .' . Fouling in heat
transfer is still a major unresolved problem (Whalley, 1 992). As the fouling process is
extremely complex, any increased understanding of fouling would be of industrial value
(Fryer et aI., 1 995).
Fouling on the evaporator heat transfer surface is a major reason for frequent and
expensive cleaning, which is required for efficient and safe operation (Fryer, 1 989).
Fouling in tubular and plate heat exchangers as well as ultra high temperature (URT)
plants has been studied considerably during the last two decades (Burton, 1 988);
however, the study of fouling in evaporation process has not received much attention.
This may be due to the extremely complex nature of the fouling process. The relatively
low temperature range involved in the evaporation process also means that the process
of deposit layer development on the surface is slow, therefore longer times should be
needed to conduct ' experimental work. No study of fouling in thin film evaporators
with rotating surface has been reported up to date.
1.5 On-farm evaporation system
The idea of milk pre-concentration in farms has attracted world-wide interest for many
years. If milk can be pre-concentrated in the farm, the pre-concentrated milk clearly
should be cheaper to transport to the processing plants than the normal milk. Similarly,
all the other volume dependent handling and processing costs should be favourably
affected by concentration. For example, pumps, milk storage capacities and refrigeration
loads would be reduced proportionately, as well as cheese vat capacities or evaporator
loads in milk powder plants. The dairy factory effluent would also be reduced.
There are three main concentration techniques which could be used in the dairy industry,
evaporation, membrane technology and freeze concentration. The freeze concentration
has not been considered for farm operation due to its high equipment costs and the
difficulty of separation of the ice particles from viscous concentrate (Dickey and Craig,
1 993). Membrane technologies, including reverse osmosis and ultra-filtration systems
have the advantage of eliminating the heat effects on the milk, but require very pure
water, which is not readily available on many farms, for cleaning. Therefore,
evaporation technology is the only viable alternative for farm pre-concentration of milk.
Robertson ( 1 987) pointed out that on-farm milk concentration using either membrane
processing or mini-evaporators may be a concept which will be of rapidly increasing
importance in the decade ahead.

The ideal evaporator to be used in on-fann evaporation system should be highly-efficient
and compact, and cause minimal damage to milk (i.e. once through evaporator). If a
suitable design is used, taking into account the whole energy consumption on the farm,
especially employing LPG fuel to drive the engine, an evaporation system would be an
effective method of on-fann concentration (Jebson et aI. , 1 993).
It has been proposed to develop a new on-fann evaporation system at the Department
of Food Technology, Massey University. The proposed system is shown in Figure 1 .2.
This system is operated as follows: The fresh whole milk is first collected in the buffer
tank and then heated and pumped into the evaporator. The concentrated milk from the
evaporator is chilled in a plate heat exchanger and then passed to the storage tank. The
storage temperature is about 5°C and the maximum storage time is about 3 days.
Finally the concentrated milk is transported by tanker to the dairy factory for further
processing and may be stored in the dairy factory for a while (maximum two days)
before processing.
Research work on both chemical and microbiological qualities of pre-concentrated milk
qualities has just been completed (Xu, 1996). The results showed that pre-concentrated
milk had an acceptable quality to be used in the industry. From a microbiological point
of view, it was found that the number of total bacteria were much less in the pre
concentrated milk than in the raw milk, therefore, the pre-concentrated milk can be
stored safely for the proposed period of time at the farm and the dairy factory.
The research work carried out in this project intended to satisfy the requirement for the
development of the on-fann evaporation system.

Plate beal exchanger
for cooling
Milk collection on farm On-Carm evaporator
Tanker S torage tank for

S ilos at processin � plant for transportation pre-concentrated milk
Figure 1.2 The proposed on-farm evaporation system

-
N
CHAPTER 2
OBJECTIVES
The thin film evaporator with rotating surface is a highly efficient and compact
evaporator, which could be a good choice for the on-farm evaporation system. Although
it has been used in producing many products, very few theoretical and systematic studies
have been undertaken on the performance of this type of evaporator. To use this
evaporator effectively and make further developments, it is necessary to understand the
evaporation process and mechanism that occurs on the rotating surface as well as its
fouling behaviour.
The objectives of this work were:
( 1 ) To develop a theoretical model that can be used to calculate liquid film thickness
and heat transfer coefficients in the rotating surface evaporator and to illustrate the
relevant factors affecting the liquid film flow and the heat transfer.
(2) To increase the understanding of the factors controlling the heat transfer coefficients
in rotating surface evaporator through an experimental study, in particular the effect of
rotating speed, cone angle, feed flow, evaporating temperature and temperature
differences on the heat transfer coefficient.
(3) To investigate the fouling behaviour on the rotating surface with changes of
operational variables, namely running time, evaporating temperature, temperature
differences, rotating speed and feed flow rate, using whey solutions as a model system.
CHAPTER 2 Objectives 14
The majority of experimental work was undertaken on the Centritherrn evaporator. A
cone evaporator with a 1 0° half cone angle and a single tube falling mm evaporator with
rotating surface were also built and used for the experimental approach.
PART II
H EAT TRANS FER IN EVAPORATORS
WITH ROTATI N G S URFAC ES

CHAPTER 3
HEAT TRANSFER IN THIN FILM EVAPORATORS
LITERATURE REVIEW
Fluid flow and heat transfer processes associated with liquid-vapour phase changes
phenomena are typically among the more complex transport circumstances encountered
in engineering applications (Carey, 1 992). The evaporation of a liquid film occurring
in a rotating system is a very complex phenomena, in which the flow pattern of the film,
the mechanisms of heat transfer and evaporation as well as the assistance of the
centrifugal force are all involved. The available information on these specific aspects
is incomplete. This literature review, therefore, focuses fIrst on the flow of films and
heat transfer associated to it, the boiling phenomena, and heat transfer enhancement by
surface rotation. Then, some related aspects of heat transfer in the evaporation process,
as wel l as the development of evaporatio n technology used in the dairy industry, are
presented. Final ly, the thin film evaporators with rotating surfaces are revi ew ed .
3.1 Liquid fUm flow
The flow of liquids in thin layers is often observed in our daily life, a common example
is the flow of rain water on window. The occurrence and applications of film flow in
modem technology are numerous and important. The falling liquid fllm under the action
of gravity can be found in thin film evaporators, film condensers, irrigated cooling
towers and falling fIlm absorbers. The thin film flow, as the result of the application
of a forced centrifugal field, can be found in the thin film evaporator with rotating
surface and rotating disk atomizers (Blass, 1979). Extensive research work has been
done to study the fllm flow on plates and tubes both theoretically and experimentally.
CHAPTER 3 Heat Transfer in Thin Film Evaporators - Literature Review 16
3.1.1 The flow regimes of the liquid film
It is well known in fluid mechanics that below a certain critical value of the Reynolds
number (Re) the flow will be mainly laminar in nature, while above this value,
turbulence plays an important part. The same is true for a film flow, though it must be
remembered that in thin films a large part of the total ftIm thickness continues to be
occupied by the relatively non turbulent "laminar sublayer" even at large flow rates
(Fulford, 1 964).
The Reynolds number for a flowing liquid ,film is defined as:
4r (3- 1 )
Re =
11
Where: r is defined as the local mass flow rate in the film per unit width of
surface (kg/m.s);
11 is the liquid viscosity (Pa.s).
Strictly speaking, the flow regime of a film cannot be defmed uniquely as laminar or
turbulent because of the presence of the free surface in film flow. Under suitable
conditions, it is possible to have smooth laminar flow, wavy laminar or turbulent flow,
where the wavy flows may be subdivided into gravity or capillary types.
In order to simplify the problem, three different flow regimes have been identified
(Fulford, 1 964; Hallstrom, 1985):
1) S mooth laminar;
2) Wavy laminar; and

3) Turbulent.
The transition from smooth to wavy laminar can occur for Reynolds numbers up to 30
(Benjamin, 1 957). Based on experiments using liquids with different physical properties
falling in vertical tubes, it appears that wave fonnation will develop when the Froude
number (defined as u2J[g.dD is greater than 1 (Jackson, 1 955).
Hallstrom ( 1 985) reported that for water, if the vapour shear is negligible, the smooth
laminar regime takes place for Reynolds numbers below 20-30, and the wavy laminar
regime starts at Reynolds numbers about 30-50.
It was found that turbulent motion first appears at Reynolds numbers between 250 and
500 (Fulford, 1 962), 200 and 600 (Blass, 1 979), or 1 ,080 (DukIer and Bergelin, 1 952).
The gradual transition to turbulent occurs with the increase of Reynolds number from
1 ,000 to 4,000 (Ueda and Tanaka, 1 975) and Reynolds numbers from 1 ,500 to 1 ,800
(Levich, 1962). Fully turbulent flow exists at Reynolds numbers between 1 ,000 to 3,000
(Hallstrom, 1 985); and Reynolds numbers greater than 10,000 (Gimbutis et ai. , 1 978).
Moresi ( 1985) suggested that although there was scatter in the published experimental
data, most investigators seemed to support a Reynolds number lower limit for the
laminar-turbulent transition between 1 ,000 to 1 ,600; with a less well-marked upper value
of about 3,200.
Schwartzberg ( 1 988) summarized work of several researchers (e.g. Yanniotis, 1983;

Chavarria, 1 983) and reported that at Reynolds number less than 500, the heat transfer
coefficient of the evaporating fluid increased linearly as Reynolds number increased, at
the range of 500 to 1 ,600, the heat transfer coefficient remained fairly constant or
decreased slightly, at Reynolds number beyond 1,600; the heat transfer coefficient
increased slightly as Reynolds number increased.
The correlations developed by Chun and Seban ( 1 97 1 ) working with water and Wilke
( 1 962) working with water and ethylene glycol show that the transition point (critical
Reynolds number) from laminar to turbulent can be detected by the change in heat
transfer rate. They, nevertheless, suggested that a Weber number (defined as [p.u2 . 8
Jcr] ll2) of the order of unity may be used as a transition criteria for falling films.
Dukler ( 1960) pointed out that the transition from laminar to turbulent flow in a thin
film cannot be expected to be sharp, since due to the thinness of films, a large portion
of the total film thickness is occupied by the laminar sublayer even at flow rates above
the critical Reynolds number. The transition is likely to be a gradual process.
3.1.2 Flow features of the falling liquid fIlms
Knowledge of falling film flow features is necessary for heat and mass-transfer
theoretical analysis (Fulford, 1964). The pioneering work in this field was that of
Nusselt ( 1 9 1 6) on the theoretical determination of velocity fields, average velocit!es and
film thicknes&sfor smooth laminar films falling at steady state on flat surfaces with zero
have
free surface interfacial shear. Since then, quite a few paper�appeared in the literature,
and most of work has been done only in the laminar flow regime, because beyond this
regime the film flow is usually so complex that no satisfactory general theory has yet
been possible (Blass, 1979). Dukler and Bergelin ( 1 952) deduced the velocity profile
and film thicknesses for a turbulent film flow based on the universal dimensionless
velocity profile equations of Nikuradse ( 1 933). Blass (1 979) obtained the falling film
thickness as a function of the Reynolds number graphically at Reynolds numbers
between 10 to 5,000, as well as the equations to calculate the mean film thickness and
the mean film velocity.
A stable film means that the film liquid completely wets the flow surface, however, even
without heat or mass transfer, dry spots may be formed as the Reynolds num�er is
reduced to a minimum value (a critical Reynolds number). Film surface instabilities are
dampened by surface tension, as established in the work of Benjamin ( 1 957), who also
stated that total stability cannot be obtained by merely increasing surface tension and
that vertically falling films are unstable at all flow rates. Brauer ( 1 956) reported
experimental values for the critical Reynolds number of films containing small quantities
of surface-active materials in solution. In this case, the value of critical Reynolds
numbers appeared to depend on the surface tension of the solution. This effect is
probably due to the layer of surface-active material present at the interface.
If temperature or concentration gradients are present in the liquid film, the Marangoni
effect should be included in the stability analysis (Blass, 1 979). The Marangoni effect
is the surface motion of the liquid as a results of local surface tension differences. If
there is a somewhat thin section of film with a smaller surface tension than the
neighbouring thick sections, then, on the basis of the tendency toward a minimum free
surface energy, the liquid in the region close to the interface is drawn from the thin
section of the film into the thicker zone. As a result, the film can be thinned so much
that it ruptures and flows into separate rivulets. This effect would be likely to occur in
highly concentrated solutions with some lower surface tension materials.
Two extensive reviews of film flow features have been given by Fulford ( 1 964) and
Blass ( 1 979).
3.2 Liquid boiling phenomena
B oiling heat transfer studies are traditionally divided in two parts: pool boiling and flow
boiling. Pool boiling represents the situation in which boiling occurs on a single heated
surface in a large pool of quiescent liquid. The flow boiling refers to the situation in
which there is bulk motion of the liquid past the heated surface, usually in tubes
(Thome, 1990).
3.2.1 Pool boiling
Pool boiling by itself -is of limited practical significance as a process, but it is studied
extensively to gain insight into the more complex problem of flow boiling. An
exhaustive review of pool boiling was given by Van Stralen and Cole ( 1 979).
The nature of the pool boiling process varies considerably depending on the conditions
at which boiling occurs.
The heating rate or the level of heat flux, the physical and thermophysical properties of
the liquid and vapour, the surface material and finish, and the physical size of the heated
surface all may have an effect on the boiling process.
Four regimes have been classically recognized for pool boiling, single-phase natural
convection, nucleateboiling, transition boiling and film boiling. These regimes are most
easily understood in terms of the so-called boiling curve: a plot of heat transfer
coefficient (hI) versus the wall superheat, given by the temperature difference between
the wall temperature (Tw) and the ambient liquid temperature (TI). The classical pool
boiling curve was determined in an early investigation conducted by Nulciyama ( 1 934).
Figure 3 . 1 shows the boiling curve for water at atmospheric pressure.
It is clearly shown on the boiling curve that at very low wall superheat levels, the heat
transfer coefficient is also low and increases slowly with wall superheat (section "a-b"
in Figure 3 . 1). This indicates that no nucleation sites may be active and heat may be
transferred from the surface to the ambient liquid by natural convection alone. The heat
transfer coefficient associated with natural convection is relatively low.
As superheat becomes large enough, nucleation at some of the nucleate sites on the
surface is initiated. This onset of nucleate boiling condition occurs at point "b" in
Figure 3.1.
Once nucleate boiling is initiated, any further increase in wall temperature causes the
heat transfer coefficients increase rapidly, which is shown as the section "b-d" of the
)
curve and corresponds to the nucleate boiling regime. The active sites are few and
widely separated at low wall superheat levels. This range of conditions, corresponding
to segment "b-e" of the curve, is sometimes referred to as the isolated bubble regime or
discrete nucleate boiling.
CHAPTER 3 Heat Transfer in Thin Film Evaporators - Literature Review . 21
Adapted from Incropera & DeWitt

(1990)
1 00000 .-----.-----;-�r___--___r_-___,
d
For water at
atmosPheric
pressure
.E 10000
1000
Nucleate
Boiling .
100 1�----5��1�0�3�2�--�1�2nO----11�00 0
Excess-temp erature , � TB (K)
Figure 3.1 The boiling curve for water at atmospheric pressure.

With increasing surface superheat, more and more sites become active, and the bubble
frequency at each site generally increases. Eventually, the active sites are spaced so
closely that bubbles from adjacent sites merge together during the final stages of growth
and release. Vapour is being produced so rapidly that bubbles merging together form
columns of vapour slugs that rise upward in the liquid pool toward its free surface. This
higher range of wall superheat, corresponding to segment "c-d" of the boiling curve, is
referred to as the regime of slugs and columns or fully developed nucleate boiling.
The increasing of the wall temperature within the slugs and columns region ultimately
results in a peaking of the heat transfer coefficient at point "d".
If the wall temperature is increased beyond this critical condition, a regime is

encountered in which the heat transfer coefficient decreases as the wall superheat
increases. This regime, which is usually referred to as the transition boiling regime
corresponds to segment "d-e" on the boiling curve.
As the wall temperature is increased, eventually a point is reached at which the surface
is hot enough to sustain a stable vapour film on the surface for an indefinite period of
time. The entire surface then becomes blanketed with a vapour film, thus making the
transition to the film boiling regime. This transition occurs at point "e" and the heat
transfer coefficient increases again after this point.
3.2.2. Flow boiling
Due to the bulk liquid flowing over the heating surface, the interaction of the flow
regime and the heat transfer mechanism make the flow boiling process become more
complex.
Two arrangements are normally considered for flow boiling inside tubes; vertical tubes
and horizonal tubes. The boiling characteristics of both orientations are significantly
different, so that different correlations are used for each of them. An extensive review
of flow boiling inside tubes was given by Collier ( 1 98 l a).
There are two important types of heat transfer mechanisms in flow boiling, nucleate
boiling and convective interface boiling. However, the proportion of these two
mechanisms varies over the length of the heating tubes.
In most (but not all) systems of practical interest, the onset of nucleate boiling is
achieved at or just beyond the point where the bulk flow reaches the saturated liquid
condition (boiling temperature) (Carey, 1 992).
Thome ( 1990) pointed out that the wall superheat required to initiate boiling in flow
boiling is not the same as that for nucleate pool boiling because the thermal boundary
layers created by the respective processes are quite different.
Many correlations have been proposed to estimate the wall superheat required for the
onset of nucleate boiling inside tubes (Bergles and Rohsenow, 1 964; Sato and
Matsullura, 1 96 1 ; and Frost and Dzakowic, 1 967). A widely quoted expression is that
of Davis and Anderson ( 1 966):
(
J
8 cr Tsar q (3-2)
!1TB =
hfg P v k
where: 1::.T
. B is the temperature difference between the wall and saturation
temperature (K),
cr is the surface tension (N/m),

Tsat is the saturation temperature (K),
q is the heat flux(W/m2 ),
htg is the latent heat of vaporization (J/leg),
P v is the vapour density (kg/m3),

k is the liquid thermal conductivity (W/m.K).
The physical properties are evaluated at the saturation conditions corresponding to the
local pressure.
3.3 Heat transfer through a thin liquid mm
In general, the heat transfer mechanisms through a thin liquid film could be conduction,
convection or nucleate boiling, the proportions of each one are mainly dependent on the
heating rate and the liquid flow pattern. The thermophysical properties of the liquid,
namely thermal conductivity, viscosity, density, specific heat, latent heat, surface tension
and boiling point elevation, and the conditions of the heating surface all may also have
an effect on the heat transfer process. As a result of the heat transfer through the liquid
film, part of liquid is evaporated. The vapour velocity over the vapour-liquid interface
may also affect the liquid fIlm flow pattern and consequently the heat transfer.
Four possible mechanisms for liquid film evaporation are discussed in the literature
(Dengler and Addoms, 1 956; Chun and Seban, 1 97 1 ; Angeletti and Moresi 1 983; and
Thome, 1990). These include:
( 1 ) Convective interface boiling, where heat transfer through the liquid film is by
conduction and convection, the evaporation occurs only at the vapour-liquid interface
of the film. The whole process is mainly governed by liquid "flow conditions and
thermophysical properties.
(2) Low rate nucleate boiling, where bubbles form and break but do not affect heat
transfer greatly. Observations show that in this case bubbles move with the fIlm without
reaching the surface.
(3) High rate nucleate boiling, where the presence of the bubbles significantly affects
liquid flow pattern and where the bubbles follow one another in channels across the
film. The mixing of the liquid entails a significant improvement in heat transfer.
(4) Vapour film evaporation, where vapour film is evolved from bubbles produced at the
tube surface.
The heating rate for a liquid film on the heating surface could be expressed as the heat
flux, which is the heat transferred per specific area, or the wall superheat, which is the
temperature difference between the wall and the liquid-vapour interface, similar to the
wall superheat used in the pool boiling curve. From the pool boiling curve for water,
it can be seen that, at atmospheric pressure, natural convection occurs when wall
superheat is less than 5°C and isolated bubble nucleate boiling occurs when wall
superheat is in the range 5 to 10°e. When a water film trickles down a tube under
gravity, the critical degree of wall superheat required for vapour bubbles to be formed
on the surface is up to 7°C, but if the heating rate is not very high, e.g. if heat flux is
less than 30 kW/m2 , only convective interface boiling applies on the water film (Billet,
1 989). Chun and Seban ( 1 97 1 ) showed in their experimental results that a wall
superheat of 3. 7°C was required for nucleation in water at atmospheric pressure.
According to Blass ( 1979), for water and aqueous solutions, transition from interface
boiling to nucleate boiling in liquid films occurs at heat fluxes in the range of 40 to 60
kW/m2 , and at temperature differences in the range of 6 to 7.SoC. In his study of the
effect of surface roughness on the boiling phenomena, Bell ( 1 982) found that the
ordinary engineering surfaces require surface superheats of the order of at least 3 to 5°C
to initiate nucleation of a stable boiling phase.
Dengler and Addoms ( 1 956) pointed out that in all cases an increase in liquid velocity
past a surface was shown to raise the temperature difference required to initiate nucleate
boiling at that surface. In other words it means that at a constant temperature difference
the heat flux contribution by nucleate boiling decreases with increasing velocity until the
boiling ceases entirely. It may be explained as that the greater velocity tends to prevent
bubble formation at the wall (O' Connor and Russell, 1 978). This effect of liquid
velocity on nucleation, incidentally, can be observed by anyone stirring a pot of boiling
water.
The wall superheat required for nucleate boiling also increases when the boiling
temperature is reduced (MUller-Steinhagen, 1 989). Thome ( 1990) argued that the

formation of the vapour core in a tube may lead to a complete suppression of bubble
growth in the liquid film at the tube wall.
In falling film evaporators, liquid distributed on the top of the tubes flows down on the
inner surface as a thin film under the action of gravity and later accelerated by the
vapour velocity. Several heat-transfer mechanisms may coexist depending on local
conditions (O'Connor and Russell, 1 978). Kroll and McCutchan ( 1 968) found the
interface evaporation plays the major role in the heat transfer at lower heat flux. But
the low rate nucleate evaporating mechanism is regarded by Sinek and Young ( 1960)
as the most probable mechanism in the falling film evaporators.
Billet ( 1 989) thought that the flow pattern in the evaporator decidedly affects heat
transfer on the liquid side of the heater, and that the amount of vapour in the liquid is
a crucial factor. He also pointed out that the calculation of the heat transfer coefficient
in evaporators is subjected to various physical laws that depend on whether boiling is
nucleate or convective.
Housova ( 1 970) presented experimental data for falling film evaporation of water in a
3 metre long tube at the evaporation temperature range of 50 to 74°C. The data
indicates that the change to nucleate boiling takes place above a wall superheat of 1 0°C.
Chen ( 1 992) found that the overall heat transfer coefficient in a single tube falling film
evaporator decreased with increasing of the temperature difference between the steam
condensing and the liquid evaporating temperatures. There was a change in the rate of
decrease of heat transfer coefficient with increases in the temperature differences
between the steam condensing and the liquid evaporating temperature at about 8K,
which was attributed to the changes of heat-transfer mechanisms in the thin liquid film
from convective interface boiling to nuc1ear boiling.
Stephan ( 1 992) stated that nucleate boiling is generally avoided In falling film
evaporators because liquid is dragged along by the vapour bubbles resulting in dry spots
forming on the wall, which can reduce the heat transfer rate and promote the fouling of
the heating surface. However, Bouman et al. ( 1 993) found that nucleate boiling starts
at temperature difference, between the liquid and wall temperature, of about 0.5°C for
milk and about 5°C for water. It also found that the transition from convective boiling
to nucleate boiling takes place at much lower heat fluxes with skim milk than with water
(van Stralen and Cole, 1 979). This can partly be explained by the difference in surface
tension, which for milk is lower (Walstra and Jenness, 1 984), but may also be due to
the presence of milk fat globules which may act as nuclei. Therefore, based on above
discussions, Mackereth ( 1995) postulated that, with water, falling film evaporators are
likely to operate solely under the convection boiling regime, while on milk, nucleate
boiling may dominate.
3.4 Effect of the adjacent gas stream on the liquid fIlm flow and heat transfer
In general, there are various correlations predicting that, as the velocity of the gas
stream is increased, the mean film thickness for a given liquid flow rate decreases for
downward concurrent flow of the two phases, due to the acceleration of the surface by
the gas stream drag (Fulford, 1 964).
The effect of the gas stream motion on the liquid film flow was first examined by
Semenov ( 1 944), he proposed an interrelationship among liquid throughput, film
thickness and tangential force which can only be applicable for a film with a thickness
not great than 1 50 microns and at moderate gas rates. A detailed analysis of the smooth
film and gas streams inside vertical tubes was also carried out by Brauer ( 1 956) .
For wavy laminar flows, an equation published by Kapitsa ( 1 948) predicts the mean film
thicknesses based on the irrigation density and mean gas velocity, when the gas stream
does not seriously affect the flow wavelength. Wavy film flow was shown to be more
stable than smooth film flow, and with a mean film thickness about 7% thinner than
smooth film at the same flow rate.
For the turbulent film regime flow, Dukler ( 1 960) provides numerical results relating the
effects of a concurrent downward flowing gas steam on film thickness and liquid
velocity. Film heat transfer under these circumstances is also determined.
Zhivaikin ( 1 962) concludes that film thickness is slightly affected by concurrent

downward flow for gas velocities up to 4 mis, but that the film thickness decreases at
higher gas velocities. He also presents a correlation to predict film thickness, for gas
velocities between 4 mls and the velocity at which spray fonnation commences.
Any pressure drop due to gas stream friction losses and the energy to create gas velocity
will cause an increase in the evaporating temperature. Consequently, since falling film
evaporators operate at low temperature difference, such pressure drop could significantly
reduce the total available heat-transfer driving temperature difference (Schwartzberg,
1988).
Zhivaikin and Volgin ( 1964) showed experimental results which indicate that pressure
drop per unit length is not always constant in the direction of gas flow. They explained
that the reason is due to changes in the shape of the gas stream velocity profile and to
changes in energy on accelerating the liquid film near the inlet. They found a
particularly interesting feature in gas flows which is that the friction pressure drop in a
wetted-wall tube is considerably larger than in the dry tube. This reason was not clear
to them. But with gas flowing down the tube, there will be a drag force on the liquid
film increasing its velocity. The energy required for this increase must be transmitted
by an increased pressure drop in the gas.
Sinek and Young ( 1 960) published an expression to calculate the friction pressure drop
in falling film evaporators and they concluded that for most cases, the vapour rate per
tube is not high enough for the friction pressure drop to deviate significantly from that
of the dry-tube value. However, the tube length employed in the commercial falling
film evaporators has been increased considerably since 1960. At the same time the
number of effects has also been increased, which results in the available heat-transfer
driving temperature difference in each effect reduced. Hence the pressure drop in the
falling film evaporators becomes significant.
lebson ( 1 988) and Jebson and Iyer ( 1 99 1 ) found that the vapour momentum is an
important and positive factor to influence the heat transfer coefficients in falling film
evaporators used in the New Zealand dairy industry. The effect of vapour velocity over
the liquid film was thought as that of an high wind blowing over a lake surface and
causing waves. This effect probably increases heat transfer especially towards the
bottom of the tube as the vapour velocity is increasing along the tube.
Carey ( 1 992) argued that the vapour flow in the core of the tube would affect the heat
transfer mechanism in the liquid film. The increasing vapour void fraction, as a result
of vaporization will accelerate the liquid flow, which produce changes in the flow
patterns. As vaporization continues, the thickness of the liquid film on the tube wall
will decrease, reducing its thermal resistance and thereby enhancing the effectiveness
of the heat convection mechanism.
3.5 Heat transfer enhancement by rotating the surface
Rotating heat transfer surface is a specific type of heat transfer intensification with
mechanical aids, which has been classified as an "active" technique (Reay, 1 99 1 ).
Within the field of intensified heat transfer, the so-called "passive" techniques (e.g.
treated surface) are routinely used to improve evaporation and condensation heat transfer
coefficients. However, the use of "active" methods, which appear to offer great potential
rewards in terms of efficiency and compactness, are less well explored and, to a greater
extent, less extensively applied. It has been pointed by lachuck and Ramshaw ( 1994)
that the use of high gravity field created by rotation is one of the most interesting and
potentially the most rewarding method among several active methods, such as stirring,
scraping or vibration.
Rotation of the heat transfer surfaces offers the following advantages (Jachuck and
Ramshaw, 1994; Yanniotis and Kolokotsa, 1996) :
( 1 ) Variable rotation speed offers a further degree of freedom on exchanger design and
operation, and the equipment volume could be reduced due to the high heat transfer
coefficient and high temperature difference that can be used;
(2) The increased "g" (acceleration due to rotating), coupled with a build-in surface
structure roughness factor, enhances the liquid film process;
(3) There is a self-cleaning action whereby rotation can better handle liquids which
contain solids as well as viscous products;
(4) Reduced fluid residence time in the heating zone will have a great influence in
processing heat sensitive liquids.
The use of centrifugal force ha�:a�monstrated to give significant degree of enhancement

in both heat and mass transfer. A disadvantage of rotating equipment is the high cost
of machining that is required. However, as automation in manufacturing is developed,
this cost could decrease (Yanniotis and Kolokotsa, 1 996).
lachuck and Ramshaw ( 1 994) have experimentally studied the convective heat transfer
characteristics of water film on four discs, each with a different surface configuration,
for various flow rates and rotational speed (250-890 rpm). Average heat transfer
coefficients as high as 1 1 kW/m2K were achieved by using a smooth disc. They
observed by using a photographic technique that there were surface waves existing in
the liquid film and claimed that the surface waves play an important role in the heat
transfer performance of the thin film on rotating discs.
The increase in the average heat transfer coefficient for increasing rotational speeds may
be due to better shear mixing, resulting from a thin film, which suggests that at higher
rotational speeds, the films get thinner and the surface waves become smaller and more
concentrated. The surface waves will cease to exist when the rotation speed increases
further.
Similar phenomena were observed by both Elsaadi ( 1992) and Lim ( 1 980). Lim ( 1 980)
explained the existence of a maximum value for the mass transfer coefficient by
suggesting that by increasing the flow rate, the film thickness increases, thereby creating
waves which induce progressively more efficient mixing in the film. However, as the
film thickness (resulting from increasing flow rate) increases beyond some optimum
value, the waves are unable to produce the higher levels of mixing required to produce
high mass transfer rates.
Convection heat transfer in axial flow through a horizontal heated tube, rotating about
its axis, has been investigated experimentally by Kuo et al. ( 1 960) with water and by
Briggs ( 1 959) with air.
Kuo et al. ( 1 960) investigated experimentally the heat transfer with axial flow of water
through a rotating horizontal annulus for the case that the liquid fills the entire annulus
and for the case that the liquid fills only half the annulus. Their results indicated that
there were at least three different flow regimes existing in the flow through a rotating
tube. Under the conditions investigated, free convection effects were found to be of
secondary importance compared to the phenomena produced by the rotation of the tube
wall. Moreover, in the range of axial flow Reynolds numbers between 60 to 1 ,000, and
rotational flow Reynolds numbers between 4,000 to 30,0000, rotational effects
predominated also over those of the axial flow. They defined axial flow Reynolds
numbers and rotational flow Reynolds numbers as Rea=Dulv and Rer=D2Q/v
respectively.
It is also worth noting that effective convection heat transfer coefficients for the
��
half-filled annulus were found to essentially equal to the values obtained with a full
channel. The reason for this is that a thin layer of liquid adheres to the rotating surface
by virtue of the rotation and is carried upward through the air. The heated liquid film
is then mixed with the cooler bulk of the main liquid stream. In this manner liquid is
heated over the entire heating area and this quasi-periodic heat flow mechanism accounts
for the large rates of heat transfer, as shown by Kuo et al. ( 1 960). The phenomenon has
been applied successfully in rotating heat exchangers (Mai, 1 95 ; Kern and Karakus,
1 958).
With water in the tube, there may exist a free fluid surface, while a gas will always fill
the entire cross-sectional area of the flow channel. The conclusion which can be drawn
from Pattenden' s work ( 1 964) is that exceptionally large overall heat transfer coefficients
can be achieved between two fluid streams separated by a rotating tube. Ferell et al.
( 1 957) also claim that a significant reduction in axial pressure loss may be achieved in
flow through a tube by increasing its rotational speed.
The details of the convection heat transfer in rotating system were given by Kreith
( 1 968) in his comprehensive review.
More recently, Yonniotis and Kolokotsa ( 1 996) reported an experimental work carried
out on a heated smooth surface of a rotating disc. They investigated the effect of
rotating speed and feed flo� rate on heat transfer coefficients by using water and corn
syrup as test liquids and boiling between 40 and 50°C. The heat transfer coefficient
increased by 3-4 times when the rotating speed increases from 0 to 104.7 rad/s, but feed
flow rate had no significant influence on the heat transfer coefficient.
3.6 Steam Condensation
The commonest medium for heating evaporators is steam. There are two mechanisms
of condensation, which are well described by the terms dropwise and film type.
3.6.1 Film Condensation
If the liquid phase fully wets a cold surface in contact with a vapour near saturation
conditions, the conversion of vapour to liquid will take the fonn of film condensation.
The condensation takes place at the interface of the liquid-vapour film covering the solid
surface. The vapour condensate travels down the surface as a continuous layer with an
increase of the film thickness under the action of gravity. Because the latent heat of
vaporization must be removed at the interface to sustain the process, the condensation
rate is directly linked to the rate at which heat is transferred across the liquid film.
The theoretical analysis for laminar film condensation on a vertical surface was first
proposed by Nusselt in 1 9 1 6. In this analysis, a number of simplifying assumptions
were made: ( 1 ) the liquid film is in laminar flow with smooth surface and its physical
properties are constant; (2) the vapour is pure and stationary; (3) the shear stress acting
at the vapour-liquid interface is negligible; (4) heat transfer across the liquid film
only by conduction.
Nusselt's theory can be expressed in tenns of a modified Nusselt number:
(3-3)
1
Nu l = 1 .47 Re 1"
Where: Nu' is defined as:
(3-4)
hL is surface heat transfer coefficient on the side of the liquid (W/m2K);

k is thennal conductivity of the liquid (WIm.K);
p is the liquid density (kg/m2) ;
g i s the gravitational acceleration (m/s2);
/-l is dynamic viscosity (Pa.s).
Since then, a substantial number of modified versions of this analysis have been
developed. A comprehensive review on the correlations for laminar and turbulent flow
film condensation on a vertical surface is given by Chun and Kim ( 1 990).
3.6.2 Dropwise condensation
Dropwise condensation is obtained only on a poorly wetted cooling surface (McCabe
and Smith, 1965). A poorly wetted surface condition can result from contamination or
coating of the surface with a substance that is poorly wetted by the liquid phase of the
surrounding vapour.
In practice, this can be achieved for steam condensation by ( 1 ) injecting a non-wetting
chemical into the vapour, which subsequently deposits on the surface; (2) introducing
a substance such as a fatty acid (e.g. oleic) or wax onto the solid surface; (3) by
permanently coating the surface with a low-surface-energy polymer or a noble metal
(Holden et aI. , 1987).
Each of these methods has its drawback and/or limitations. The effects of the first two
methods are generally temporary, since the resulting surface films eventually are
dissolved and washed off from condenser surface under common industrial operation
conditions. The third method of promoting dropwise condensation is of particular
interest because it holds the prospect of providing continuous dropwise condensation,
but generally considerable cost will be added to the fabrication (Carey 1 992).
Dropwise condensation is generally the preferred mode of condensation because the
resulting heat transfer coefficient may be as much as an order of magnitude higher than
that for film condensation under comparable circumstances. Billet ( 1989) stated that the
heat transfer coefficients in dropwise condensation are higher than those encountered in
purely film condensation by a factor of about 6-20.

During dropwise condensation, the condensate is usually observed to appear in the form
of droplets, which grow on the surface and coalesce with adjacent droplets. When
droplets become large enough, they are generally removed from the surface by the
action of gravity or drag forces resulting from the motion of the surrounding gas. As
the drops roll or fall from the surface, they merge with droplets in their path, effectively
sweeping the surface clean of droplets. Droplets then begin to grow on the · freshly
exposed solid surface. This sweeping is renewal of the droplet growth process, which
is responsible mainly for the high heat transfer coefficients associated with dropwise
condensation.
Despite numerous studies of dropwise condensation over the years, its mechanism
remains the subject of debate (Carey, 1992). Two different types of models have been
proposed:
( 1 ) The first model type is based on the premise that droplet formation is a
heterogeneous nucleation process. Droplet embryos are postulated to form and grow at
nucleation sites, while portions of the surface between the growing droplets remain dry.
This type of model apparently was first proposed by Eucken in 1 937.
(2) The second type of dropwise condensation model postulates that condensation occurs
initially in a filmwise manner, forming an extremely thin film on the solid surface. As
condensation continues, this film eventually reaches a critical thickness, estimated to be
about 1 Ilm, at which point it ruptures and forms droplets. Condensation then continues
on the surface between the droplets that form when the film ruptures. Condensate
produced in these regions is drawn to adjacent drops by surface-tension effects.
Droplets also grow by direct condensation on the droplet surfaces themselves. This
second model was proposed by Jakob in 1936. Modified versions of this model have
also been proposed by Kast ( 1 963) and Silver ( 1 964).
B oth dropwise condensation models are supported by several experimental
investigations. Using an optical technique to indicate changes in the thickness of very

thin liquid films, Umur and Griffith ( 1 965) found that, at least for low temperature
differences, the area between droplets was, in fact, dry. Their results indicate that no
film greater than a monolayer existed between the droplets, and that no condensation
took place in those areas. However, the results presented by Welch and Westwater
( 1 96 1 ) and Sugawara and Katusuta ( 1966) indicate that condensation occurs entirely
between droplets on a very thin liquid film.
As suggested by Collier ( 1 98 1 b), it may be that droplet nucleation dominates at low
condensation rates, with the film disruption mechanism taking over at higher
condensation rates.
A detailed discussion of theoretical aspects of dropwise condensation can also be found

in a series of publications by Tanaka ( 1 975, 1979, 1 98 1) .
Other methods of promoting dropwise condensation has been examined and reported,
such as, surface vibration (Brodov et ai., 1977) surface rotation (Nocol and Gacesa,
1 970; Sparrow and Gregg, 1 959), electrostatic fields (O'Neill and Westwater, 1 9 84).
3.6.3 Condensation in the presence of non-condensable gases
The most serious problem in steam or vapour condensation is the presence of non
condensable gases. Small amounts of non-condensing gases usually are dissolved or
occluded in the liquid food being concentrated. Air also leaks into the vacuum system
through j oints and connections. It mixes with the steam or vapour flowing towards the
condensing surface, and unless a means of steadily removing it is provided, it will
accumulate at the condensing surface impeding the process of condensation, and
impairing the heat transfer coefficient.
Air is heavier than steam and would be expected to collect at the bottom of the
apparatus. However, diffusion and steam flows due to condensation ensure that the air
remains mixed with the steam and it was considered that there would be more air in
dead comers where the flow is stagnate (Lyle, 1 947).
As only the vapour is condensed, the concentration of the non-condensable gases at the
condensing surface is higher. This, in turn, decreases the partial pressure of the vapour
at the condensing surface and the corresponding saturation temperature at the condensing
surface is lower than the bulk temperature. Therefore, the non-condensable gases reduce
heat transfer coefficients in two different ways: (a) they form a layer on the condensing
surface (air blanketing effect) and (b) non-condensable gases lower the condensation
temperature at a given total pressure affecting the heat transfer efficiency.
Only traces of air could cause a steep decrease, as any concentration of air greater than
0.01 % can have a drastic effect on the condensing heat transfer coefficient (Mincowycz
and Sparrow; 1966, Sparrow et al. , 1 967). Condensation rates are reduced by 1 0% ,
when flowing steam contains 2% air, whereas the reduction i s 70% in stagnant zones
(Collier, 1 98 1 a).
In order to maintain good heat transfer performance, non-condensable gases removal is

usually accomplished by vacuum pumps or ejectors connected to the condensing vapour
chamber. It is important to set the proper de-aeration rate, because the de-aeration rate
determines the steam velocity at steam inlet and the average concentration of non
condensable gases at the condensing surface. Mackereth ( 1 993) shown that at low de
aeration rates, the average concentration of non-condensable gases at the condensing
surface increased. The heat transfer coefficient dropped slowly to approximately half
the original value in a couple of hours in a single tube evaporator when the de-aeration
line blocked. However, the removal of non-condensable gases reduces thermal
efficiency. Schwartzberg (1988) pointed out that the mix of gas and vapour venting
from a vacuum device often contain 97% to 98% vapour, which is wasted. Steam losses
might be reduced if we were better able to control levels of non-condensable gases.
CHAPTER 3 H eat Transfer in Thin Film Evaporators - Literature Review 38
3.7 The development of evaporation technology in the dairy industry
The evaporation of milk has been known for many years, even as early as in the year
1 200 A.D. when Marco Polo described the production of a pasta, which was effectively
a sort of milk concentrate, in Mongolia (Westergard, 1983).
The most simple evaporator was an ordinary open pan heated with fire, hot water or
steam. The simplest evaporator was used to produce storable milk with adding sugar
by De Heine in 1 8 1 0 (Wiegand, 1985).
In 1 8 1 3, Howard invented a vacuum pan with a condenser to deal with the vapour
produced (Taylor, 1 982). This invention enabled the lowering of the evaporation
temperature, improved the quality of products and was an important step on the way to
industrial production of concentrated and durable liquid dairy products.
In 1 850, Gail Borden was the first person to concentrate and sell milk on an industrial
scale. He used a vacuum pan with a heating coil to enlarge the heating surface
(Arrnerding, 1 966). At this time sweetened condensed milk was produced in the United
States.
In 1 852, Robert Von Seelouitz was granted a patent for an evaporator, the so-called
Robert evaporator, which had a vertical tube bundle arranged around a wide central
circulation tube. A natural circulation, which improved the heat transfer, occurred in
it. This kind of evaporator was improved and used in the dairy industry for quite a long
time. Its use was animportant precondition for high capacities and continuous operation.
In order to cope with highly concentrated solutions, the vertical tube bundle was
separated from the separator, which made it possible to pump the concentrated milk to
be cycled in the evaporator. This kind of evaporator is called a forced circulation
evaporator.
In 1 90 1 , Kestner invented the so-called climbing-film evaporator, which was fitted

with longer heating tube (about 7 m long and 50 mm diameter) and arranged with a
chamber for the separation of the liquid from vapour alongside the calandria. This type
of evaporator was developed on the basis of the Robert evaporator, but heat transfer
coefficients were higher and was operated with less recycle than the Robert evaporator.
In the climbing film evaporator, the amount of recycle is controlled, but not in the
Robert evaporator.
With the development of production, high quality products were required. The problems
in using a climbing film evaporator in dairy industry were shortly realised, because the
hold up of the liquid in evaporator was high, that means the residence time of the milk
in the evaporator was rather long. The high static pressure which prevents the liquid
from boiling in the bottom parts of the tube, worsens the heat transfer in this area
considerably. The evaporator cannot be used effectively with small temperature
differences.
The disadvantages of the climbing film evaporator were recognized in the earlier 1 930' s.
In 1 935, the falling film concept was introduced by D. D. Peebles and P. D. V. Manning
and the first patents were in this period (Hallstrom, 1 988a).
However, because stainless steel was very expensive at that time, no suitable pumps
were available and the problem of how to distribute the liquid evenly onto the heating
tube was not resolved, a commercial falling film evaporator was not manufactured for
use in industry until 1 953 by Wiegand in Germany. Since then, falling film evaporator
has been developed quickly. In many applications this type of evaporators have almost
entirely displaced the other types of evaporators and this is true in the case of the dairy
industry. Nowadays, all evaporators used in New Zealand dairy industry are falling film
evaporators (FFE).
During the development of evaporating technology, two methods of reducing energy

consumption were created and are still used today. One is multiple effect evaporation
and another is vapour recompression.
In England in the year 1 825, and in France in the year 1 833, it was suggested to use the
heat contained in the vapour produced in one evaporator for the heating of another
evaporator and, if need be, to heat a third evaporator by means of the vapour produced
in the second one, i.e. to operate according to the so-called multiple effect principle
(Hallstrom, 1988b). But, before about 1 9 1 0, most evaporation plants had only one
effect. The first continuous mUltiple effect evaporator was used in the middle of 1 8th
century in the sugar industry (Hahn, 1 985). The multiple effect operation was
introduced into dairy plants at the end of the nineteen thirties in Europe for producing
evaporated cream (Wiegand, 1 985). The theoretical specific steam consumption, in
terms of the percentage of kg steam used per kg water evaporated, equal to the
reciprocal value of the number of effects, it is therefore 1 00% with only one effect, 50%
with two effects, 33.3% with three effects. The number of effects, however, is limited
by the total temperature difference available. In the food industry it is often necessary
to avoid boiling temperatures above 70 DC when dealing with sensitive products due to
the formation of deposits on the heating surface and the effects on product quality. On
the other hand it is very costly to operate with boiling temperature below 40 °C. Low
temperatures not only result in a large increase in cooling water consumption, but also
in a considerable increase of the evaporator size, because of the higher vacuum required,
consequent larger vapour volumes and higher viscosity products with lower heat transfer.
Nevertheless, the number of effects in an evaporator has been steadily increased and six
and seven effects evaporators are now used.
In vapour recompression, the vapour can be compressed thermally by high-pressure

steam-jet ejectors; or mechanically, by positive-displacement or centrifugal compressors
as well as fans, to higher pressure and used as heating medium. It is obvious that a
large saving in energy is obtained by compressing the vapour because the enthalpy
difference between the vapour to be compressed and steam is about 24 kJ/kg, while
between water and steam about 2240 kJ/kg. Although steam jet compressors were
equipped on milk evaporators in Europe: after WorId War I, and mechanical vapour
compressors were first found in the Swiss dairy industry during World war II (Wiegand,
1 985), the oil crisis in 1 973 greatly stimulated their use (Hallstrom, 198 8b). In general,
compressors can raise the saturation temperature of the vapour by 15-20°C, while fans
can raise the saturation temperature by 4-6°C. Fans are now preferred, instead of
compressors, because of their simplicity in design, ease of operation, and therefore lower
cost (Mackereth, 1 995).
After the two energy crises of 1 973 and 1 978, energy requirement became a serious
problem, the number of effects increases greatly. In 1 966, the first quadruple effect
evaporation plant was produced, in 1974, the first five effect plant, in 1 976, the first
six effect plant and shortly after this, the first seven effect evaporator was installed
(Hahn, 1 985). In 1 979, the mechanical vapour recompression was used once again in
Germany in a triple effect plant for the evaporation of whey (Kessler, 1 985).
The reasons for the widely use of falling film evaporator in the dairy industry are the
peculiar characteristics of this evaporator, which can be described as follows:
( 1 ) FFE can be operated at any low temperature difference. This means that FFE can
be theoretically equipped with as many effects as one wanted;
(2) The lower temperature difference that can be used in FFE is also favourable to
multiple effects operation and mechanical vapour recompression by fans;
(3) The very short residence time of the product in the evaporator is a result of the small
amount of liquid hold up and the high flow velocity. Since the product is generally
processed in single pass all liquid particles have an almost equal time of direct contact
in the evaporator. Thus undesirable microbiological, physical and chemical changes in
the milk resulting from excessive heat treatment are avoided;
(4) Due to the relatively high flow velocity and low temperature difference, the deposit
on the tube wall is small, i.e. the operating time before cleaning can be largely extended;
CHAPTER 3 Heat Transfer in Thin Film Evaporators · Literature Review 42
(5) High final concentrations are possible in FFE since the high vapour velocity propels
viscous products without plugging, and the low temperature difference used makes it
possible to reach high concentration without scaling of the tubes;
(6) The capability of evaporation of FFE can be extended to satisfy any requirement due
to single construction and the possibility of using longer tubes ( 1 5 m or more). Low
operating costs, low steam consumption, and high quality can be achieved more easily
in bigger evaporating plants (Wiegand, 1 978);
(7) A relatively very small amount of cleaning agents is needed as the liquid volume in
the evaporator is low.
It can therefore be seen that falling film evaporator fully satisfies the demands of the
modem dairy industry of today, i.e. the possibility of large capacities, economical
running, reliable and continuous operation, careful treatment of the products.
Evaporation is the only process used for the concentration of milk in New Zealand
(Mackereth, 1 995). The potential rival processes, which can also be used to remove
water from milk, are reverse osmosis and freeze concentration. Reverse osmosis may
become competitive with evaporation if maximum processing temperatures of 40-50°C
are permitted, and only a low concentration product is required. It was reported that
reverse osmosis can concentrate skim milk only up to 25% total solids, and the final
concentration required must then be obtained by subsequent evaporation (Carie, 1 994).
With rapid decrease in capital cost and increase in membrane life, reverse osmosis is
now been used as a first stage of whey concentration in whey powder manufacture
(Bennet, 1996). Freeze concentration is only likely to be economic for low volume
products that are hypersensitive to heat. Therefore, evaporation technology will remain
an important process in the dairy industry, especially for the milk powder manufacturing
process, for the foreseeable future (Mackereth, 1995) .
It is well known that the efficiency of the evaporator i s much higher than the drier.
Therefore, concentrates with higher final total solids produced in the evaporator to the
drier means that the total energy consumption in the whole process of milk powder
production will be reduced. In the present milk powder manufacturing process, the
maximum allowable total solids of milk concentrates are mainly limited by the drier
(Mackereth, 1 995). If a breakthrough allows a higher viscosity feed to the drier, the
falling film evaporator would soon limit the total solids of the concentrates. B y then,
an evaporator, which could produce a further higher total solids concentrate, will be
needed. Thin film evaporators with rotating surfaces may will be chosen for this
additional concentration (Bouman and Waalewijn, 1994a, b).
3.8 Thin film evaporators with rotating surfaces
It has already been recognized that the centrifugal force generated in a rotating system
may be utilized to enhance the heat transfer performance. One of the applications of
this phenomena is in the thin film evaporator with rotating heaters, which has been used
for decades. From the aspect of modern process engineering, they are an interesting
development and can be considered more as a welcome companion than as competitor
to other types with short residence times, although the fields of application may overlap.
3.8.1 The development of this type of evaporator
For many years, it was believed that the concentration of heat sensitive materials should
be carried out under very low temperature conditions to avoid damage. On this
principle a number of evaporators were installed which were designed to boil liquid at
temperature as low as 2 1°C (Gray, 19 8 1 ). Unfortunately, this is very uneconomical and
involves considerable practical difficulties with cooling water. Now, it has been shown
that in many cases it is the time temperature integral which controls the extent of
deterioration, and that it is possible to evaporate quite sensitive materials at relatively
higher temperatures than had previously been thought, provided the length of time for
which the product is maintained at this temperatures is sufficiently short. Therefore,
much attention has been paid to reducing the heat contact times in evaporators to the
very minimum (Leniger, 1 986).
Experience gained in industrial practice has revealed that thin film evaporators with
rotating wipers are by no means successful in all cases where heat unstable products
have to be evaporated, because even residence times of about a minute are often still too
long. Thus, if a residence time of a mere second were to be specified for a liquid with
a viscosity as high as 20 Pa.s or more, no evaporators would be available. The
mechanical and process engineering problems involved in problems of this nature have
been approached from an entirely different angle, and a new principle has been devised
to cope with them. Thus a design has been developed that permits evaporation under
extremely mild conditions; for example, the residence time in the actual evaporation
zone may be only a fraction of a second (Billet, 1989).
Rotating the surface on which evaporation takes place appears to be a logical way of
developing this type of evaporator. The rotating conical heating surface was first
developed in the US by Hickman, for sea-water distillation, beginning in 1 935. The
design was re-discovered and modified by Mautner in Yugoslavia and was presented as
a commercial unit by Alfa-Laval in Sweden in 1959 (Hallstrom, 1 969). In 1 980's, a
similar type of evaporator was built in Germany and Japan.
Depending on the orientation of the rotating cone, there are three different designs of
this type evaporator available in the market. The most successful one is the Centritherm
evaporator, which has a cone rotating about its vertical axis. There are three sizes:
CT 1B, CT6 and CT9. Their heating surfaces are 0. 1m2, 2.4m2 and 7 . 1 m2 respectively.
For CT6 and CT9, there are multi-cones stacked upon each other. The largest capacity
of Centritherm evaporator was reported as about 5,000 kg/h water evaporation
(Mannheim and Passy, 1 974).
The second one, called Liprotherm, has an inverted rotating cone. This design has a
compact form as it includes the pre-heater and condenser in a same unit. A sight glass
is also placed on the unit which can be used to observe the evaporating liquid film flow
(Billet, 1989).
A recently developed model called Evapo has a horizontal rotating cone. There are six
different available sizes for this evaporator. The heating surfaces are in the range of 0. 1
to 5m2, with water evaporation rates from 70 to 2,000 kglh. The whole machine is
automatically controlled and has a sight glass (Anon, 1 990).
Figure 3.2 shows the different designs of this type of evaporator.
3.8.2 The features of the rotating surface evaporator
This type of evaporator offers the following features:
( 1 ) It has an ultra short residence time, which is important for heat sensitive products
such as enzymes, antibiotics, herbal medicines, protein solutions, fermented liquids, etc ..
When evaporating or concentrating these heat-sensitive liquids, it is required that the
process liquid receives a specific quantity of heat at a minimum temperature for an
extremely short time period. The rotating surface evaporator satisfies all three of these
important criteria. The liquid film produced by centrifugal force is a scant 0. 1 mm
thick, which is thinner than a mechanically spread film.
(2) It operates with high concentrated liquids. It was reported that the milk concentrate
produced in the Evapor unit can be up to 85% dry matter (Anon, 1 990). However it is
unlikely for skim milk to be concentrated to such high level as gelling would occur
when concentrate total solids is beyond 65% (Wood, 1 982). So it would p�obably be
used for sweetened milk concentrate (i.e. sugar may be added into the products).
(3) The centrifugal force generated by the rotating cone inhibits any tendency for the
process liquid to foam, consequently improving efficiency. The application of
centrifugal force to the liquid on the moving heat transfer surface, presses the liquid
onto this surface and thereby eliminates foaming in a way not possible with other type
CHAPTER 3 Heat Transfer in Thin Film Evaporators ... Literature Review 46
PF
t�
'--"'!J----''i'-t'''PI V
-----1-++ -
Cooling
S
T water
out
Cooling
tc
(1) Centritherm (2) Liprotherm
F - Feed
Sight V ... Vapour
glass S - Steam
, p ... Product (concentrate)
C ... Condensate
(3) Evapo
Figure 3.2 The different designs of the film evaporators with rotating surfaces
of systems that use fixed heat transfer surfaces (e.g. downflow systems, mechanically
agitated systems, etc.).
(4) It has a high overall heat transfer coefficient. Rotation of the heating surface does
not only improve the heat transfer on the liquid side by effectively distributing and
quickly passing the liquid to be evaporated on the surface, but also benefits the steam
condensation on the other side. As soon as the steam condenses on the outer surface
of the cone, the condensate forms droplets, which are thrown off the rotating surface by
centrifugal force. Therefore, no condensate film, which would offer resistance to heat
transfer, is formed. The dropwise condensation is considered to be a major mechanism
of condensation, which gives rise to very high heat transfer coefficients (Mannheim and
Passy, 1 974). Further, the fast moving liquid film also reduces the formation of deposit
on the heating surface because the formation of deposit on the heating surface is
controlled not only by the temperature but also by the movement of liquid on the
heating surface (Wiegand, 1 978).
(5) This type of evaporator is also highly suitable to concentrate high viscosity products.
In conventional falling film evaporator the liquid film flow is assisted by vapour
velocity, but at high liquid viscosities this is less effective (Jebson and Iyer, 1 99 1 ) .
Hence, the final concentration o f the products is limited. In the Centritherm evaporator,
however, the transporting force can be greatly increased with mechanical assistance,
which means considerably higher velocity and hence larger heat transfer coefficients and
the possibility of obtaining a higher final concentration. The centrifugal force, generated
in the rotating system at high rotational speeds, could be more than one hundred times
greater than the force of gravity (Hallstrom, 1969). In addition, high viscosity liquids
tend to form runnels which do not to wet the heating surfaces properly, but rotating the
surface could overcome this problem.
(6) This type of evaporator meets the equipment for 'clean' production plants, since the
amount of volatile decomposition products, which can enter the atmosphere or the
effluent through the vacuum system, is reduced due to extremely short residence times.
In view of the increasing severity of legislation on the environment, this fact may be a
convincing arguement in swaying decision on equipment selection in favour of these

evaporators (Billet, 1 989).
In this type of evaporator, energy must be supplied during start up to accelerate the
rotating cone from unmoving to the angular velocity (n). Energy is also required in the
rotor to give the liquid its final velocity and overcome the frictional losses. It was
estimated that the electric energy in this type of evaporator is between 1 and 1 0 % of
thermal energy input (Yanniotis and Kolokotsa, 1996).
3.8.3 Industrial use of rotating surface evaporators
A literature search for the period 1 969- 1 995 showed that there were 30 published papers
dealing with the application of Centritherm evaporators. The Centritherm evaporator can
be used in producing of a wide range of products, such as: fruit juices (Hallstrom,
1969), coffee and tea extracts, egg, yeast extracts (Pajunen et aI., 1 974), milk products
(Bouman and Waalewijn, 1994a,b), gelatin, meat extract (Traegardh, 1 974), malt extract
and brewery syrups, low-alcohol and alcohol-free beers (Oliver-Daumen, 1 982, 1 986),
glucose products (Shinn, 197 1), antibiotics, herbal medicines and enzymes (Anon, 1 992).
Clarified apple juice containing 9- 10 % dry matter was concentrated in a Centritherm

evaporator. The feed flow (600 to 1 ,400 lJh), steam temperature (90 to 1 1 5°C) and
evaporating temperature (45 to 60°C) were varied. It was conduded that the feed flow
had a much greater effect on the degree of concentration than the other two variables
(Nong, 1969) .
Lewicki and Kowalczyk ( 1 979) conducted experimental work using a C entritherm

evaporator with apple juices ( 10, 20 and 30% dry matter, DM) and sweet whey (6 and
1 3% DM) by varying the temperature differences (20, 30, 40, 50K), evaporating
temperatures (30 to 40°C), and the angular velocities of the rotating cone (72, 1 08 and
144 rad/s). The Reynolds numbers averaged for the feed and concentrate streams were
calculated to be in a range 2 1 to 2 1 3 . They found that the overall heat transfer

coefficients was independent of the type of food, its solids contents and the angular
velocity and depended only on Reynolds number and temperature difference. It was
also found that the residence time of material in the evaporator was 0.6 to 1 .6 seconds
and was longer for higher temperature difference at constant Reynolds number and
inversely proportional to the Reynolds number at constant temperature differences.
CHAPTER 4
EXPERIMENTAL METHODS FOR DETERMINING
HEAT TRANSFER COEFFICIENT
4.1 Objectives
The objectives of this experimental work were to assemble and operate a single-tube
falling film evaporator with rotating surface and a cone evaporator with 1 0° top half
angle and also to evaluate the effects of operating variables on the overall heat transfer
coefficients in these two evaporators as well as the Centritherm evaporator.
4.2 Materials
4.2.1 Water
Tap water was used because it is a liquid with well known physical properties.
4.2.2 Sugar solution
Sucrose used was Chelsea white sugar produced by the New Zealand Sugar Company.
The reasons for using sugar solution are that it is cheap, non-corrosive, not heat
sensitive (in the temperature range of 50 to 70°C) and easily available. Its physical
properties are also reasonably well known. The sucrose granules were dissolved in hot
water to make up the desired sugar solutions.
4.2.3 Reconstituted skim milk
Low heat skim milk powder, manufactured by the Tui Milk Products Company, was
used to make up the reconstituted skim milk solutions. The reason for using
reconstituted skim milk instead of fresh milk is that the seasonal and compositional
variation of milk is avoided.
Chapter 4 Experimental Methods for Determining Heat Transfer Coefficient 51
4.2.4 Chemicals used for Cleaning-in-place (CIP)
1 ) Caustic alkaline solution (AC- 1 80): caustic soda, made up to a strength of 1 %;

2) Acid solution (AC-300): nitric acid and phosphoric acid, made up to a strength
of 1 % ;
3) High strength sanitiser: Klenz Iodophor, made up to a strength of 1 .5%.
All these chemicals were supplied by Economics Laboratory, Hamilton, New Zealand.
4.3 Equipment
The equipment used in the experimental work is listed below:

( 1 ) A Centritherm evaporator (CT IB-2, ALFA-LAVAL, Lund, Sweden)
(2) A falling film evaporator with rotating surface, specially built at the Department of
Food Technology;
(3) A cone evaporator with 1 0° top half angle, specially built at the Department of food
Technology;
(4) A temperature recorder (32 channel recorder/datalogger, MOLYTEK, Japan);
(5) A refractometer (ATAGO, No.25870, Japan) ;
(6) A drying oven (Clayson, Catalogue 01 1 24, Laboratory apparatus Ltd, Upper Hurt,
New Zealand);
(7) Other instruments used include an analytical balance, a measuring cylinder and a
stop watch.
(8) Brookfild viscometer was used for measurement of skim milk viscosity.
4.3.1 Centritherm evaporator
The whole apparatus of Centritherm evaporator is shown in Figure 4. 1 . Figure 4.2

shows the schematic diagram of the Centritheim evaporator system.
Chapter 4 Experimental Methods for Determ i n i ng H eat Transfer Coefficient S2
F i gure 4. 1 The whole a p para tus o f th e Cen tri th erm evaporator a n d t h e fal l i n g
fil m e v a po rator w i th a rotati n g tube
1 . Feed container 7. Steam regulator
2. Feed pump 8. Thermometer
3. Rotameter 9. Feed tube
-' 4. Preheater 10. Concentrate rube
5. S team-water mixer 1 1 . S team j acket
( �) 6. Pressure regulator 1 2. Steam trapper
13. Concentrate pump
14. Concentrate container
- . 1 5 . Condenser
s::
Cooling water out 1 6. Condensate pump (l)
1 7 . Condensate container ::r

o
Cl.
---...:- 1 8. Vacuum unit !Jl
�;;;;;;;� Cooling water in • Thermocouple probe 0'

-.
CJ
(l)
Cold CD
water 3
::r
I II 5·
to
:r:
(l)
�
--f
II iU
Cold water ::l
!Jl
CD
...,
Figure 4.2 Schematic diagram of the Centritherm evaporator system

4.3.2 Single-tube falling film evaporator with rotating surface
A pilot scale single-tube falling film evaporator with rotating surface was linked to the
Centritherm evaporator system. Thus, with the exception of the heating column, vapour
chamber and temperature measurement system, all other parts of the apparatus were the
same as those of Centritherm equipment, i.e. feed pump, concentrate pump, condensate
pump, preheater, condenser, vacuum unit, and steam regulation system. Figure 4.3
shows the single-tube falling film evaporator with a rotating surface.
For this evaporator, the heating column contains a 32.00 mm O.D., 1 .6 mm thick, 2 m
length heating tube, which was coaxially placed inside a 73 mm LD., 2 m length tube,
forming a steam jacket between these two tubes. The total heating area (referring to
outside surface) was 0.206 m2• Because of the height of the pilot plant ceiling, the
heating tube height of the falling film evaporator was limited to 2 m.
A 20 mm O.D., 0.75 m length tube (feed tube) was centrally inserted into the heating
tube. The feed tube was fitted at the end with a distributor in which the gap between
the tube and the plugger can be adjusted. The structure of the feed tube is shown in
Figure 4.4.
A variable speed motor (LS80L- l , 0.75kW) was mounted on the top of evaporator. The
heating tube was rotated through a driving chain.
Due to the rotation of the heating tube, the steam jacket required a bearing and a seal.
After several trials, a stainless steel house which contained a seal was installed and a
teflon strip of 0.7 mm thickness and 20 mm wide was used as a bearing.
A cyclone separator was mounted in the vapour chamber to separate liquid particles
entrained in the vapour.
()
=r
1 . Feed container 1 1 . S team trapPer III
(7 ) "0
2. Feed pump 12. Separator ro
...,
3 . Rotameter 13. Concentrate pump �
19 4. Preheater 14. Concentrate container m
x
5. S team-water mixer 15. Condenser "0
8 (l)
...,
7 6. Pressure regulator 1 6. Condensate pwnp �r
7. S team regulator 17. Condensate container (l)
::J
8. Thermometer 18. Vacuum unit
S team �
9. Feed tube 19. Driving motor �
/ 10 1 0. S team jacket Tbennocouple probe (l)
:T
•
/ 0
C.
6 IS (J)
0
...,
Cold
,
Cooling water out
- 0
(l)
ro
I f) ...,
3
Cooling water in s·
s·
4 10
I f) I
(l)
P) �
-i
...,
III
::J
(J)
,..-, 1 2
,3
bd iii
...,
18 ()
Cold water 0
(l)
::::
o·
1 iii'
2.
Figure 4.3 Schematic of the si ngle-tube falling film evaporator with a rotatin g surface
1 2
1 . Thermocouple probe
2. Valve
3 . Fixer
4. Adjustable distributor
Figure 4.4 Schematic diagram of the feed tube

C hapter 4 Experimental Methods for Determining Heat Transfer Coefficient 57
4.3.3 Cone evaporator
The cone evaporator, made up of stainless steel, had a rotating cone with a 1 0° top half
angle. The surface area for heat transfer in this evaporator was 0. lm2 . The details of
the dimensions of the cone evaporator are shown in Figure 4.5.
Similarly to the single tube falling film evaporator, the cone evaporator was also linked
to the Centritherm evaporator system.
4.4 The evaporation process in thin fIlm evaporators
The liquid to be evaporated was pumped by a variable speed feed pump (mono pump
type 1 ) through a Spiraflo heat exchanger (Model TT 0.75/0.5-3), where the liquid was
heated up by hot water to evaporating temperature, and then fed into the feed tube.
After passing through the feed tube, the liquid flowed down the inside of the heating
surface (cone or tube) as a thin evaporating film. The heat transfers from the inside of
the heating surface to evaporate a portion of the liquid film. The unvaporized liquid and
vapour flowed straight to the bottom of the vapour chamber where the unvaporized
liquid (concentrate) was pumped out by a concentrate pump (mono pump type 1 ) and
collected in a bucket. The vapour passed through a spiral plate condenser, where the
vapour was condensed. The condensate was pumped out by a centrifugal pump (type
OM) and was collected in a bucket for flow rate determination. Cold tap water flowing
inside the spiral tubes of the condenser provided the heat sink for the condensation.
Steam was fed into the steam jacket. The heat flux, which was released by condensing
steam outside of the heating cone or tube, was transferred to the liquid to be evaporated
by conduction and convection. The steam condensate flowed down the outside of the
heating cone or tube and was removed by means of an ejector into the vacuum unit
located at the bottom of the steam jacket.
F P
�t Chain
S -
Driving motor
F - Feed
P Product (concentrate)
-
V - Vapour
C - Condensate
S - S team
C V
� = 10°
d = 80 mm
D = 256 mm
1 = 507 nun
Figure 4.5 Schematic diagram of cone evaporator

and the dimensions of the cone
The hot water used to heat up liquid in the Spiraflo heat exchanger was obtained by the
use of a hot water mixer into which steam and cold tap water were fed. The pressures
of steam and cold water were controlled by means of a steam regulator valve and a
pressure regulator valve respectively.
Vacuum was provided by the Centritherm vacuum unit. This unit consists of a container
for water, a circulation pump and three ejectors with built-in check valves. When the
water was pumped by the circulation pump through the ejectors, a vacuum was created.
One of the ejectors was used to remove steam condensate from the steam jacket. The
other two were used to remove non-condensables from the evaporation side.
The vacuum was adjusted manually during operation, and released at the end of the
operation, by means of a vent cock on the venting line.
The heating section and vapour stream tube lines were assembled with removable
clamps and rubber gaskets. The liquid flow lines were made from transparent plastic
hose. The system was carefully sealed and checked for air leakage during each run.
4.5 Instrumentation
4.5.1 Steam regulator
Steam pressure was controlled by means of a steam regulating valve (type SRV-5) with
a temperature regulator. The adjustment of the steam regulating valve as well as the
temperature regulator and more detailed information can be found in the ALFA-LAVAL
Centritherm instruction book ( 1 985).
4.5.2 Flowmeter
The inlet feed flow rate was measured by means of a stainless steel variable area
rotameter with a capacity of 0.4-4.5 litres per minute of water at 20°C. The rotameter
was calibrated by discharge weight versus time measurements at the operating
temperature.
4.5.3 Temperature measurement
Temperatures were measured by type J thennocouples (iron/constantan), which were

connected to a recorder (32 channel recorder/datalogger, MOLYTEK). The
thennocouple probes were monitored at nine different points, i.e. No. 1 measured the
steam temperature after regulating; No. 2 measured the steam temperature entering the
steam jacket; No. 3 measured the steam condensate temperature; No. 4 measured the hot
water temperature. No. 5 and 6 measured the feed temperature at inlet and outlet
preheater; No. 7 measured the feed temperature just before entering the feed tube; No.
8 measured the concentrate temperature; No. 9 measured the vapour temperature. All
these measurement points are indicated in Figure 4.2. Before starting the experiments,
all thennocouples were calibrated in a constant temperature bath at boiling and freezing
points, and intennediate points against a standard mercury thennometer. The recorder
plotted all measured temperature on a chart at 1 0 second intervals.
The thermocouples were calibrated in a water-ice mixture and in boiling water using an
certified mercury-in glass thermometer.
4.5.4 Pressure measurement
A vacuum gauge was used to monitor vapour line. Two steam pressure gauges were
used to monitor steam line pressure for the heating column and the water mixer
respectively. A water pressure gauge was used to monitor water line pressure to the
water mixer. The gauge positions are shown in Figure 4.2.
4.5.5 Rotating speed measurement
A digital tachometer (made in USA) was used to measure the rotating speed .._
The details of using the tachometer can be found in its operation manual.
4.6 The variables and their ranges
The following variables were selected for this experimental work:

a) Overall temperature difference
b) Evaporating temperature
c) Rotating speed
d) Feed flow rate
e) Cone angle
After the apparatus was assembled, considerable time was taken to calibrate temperature
measurements. Because temperature differences were used to calculate the heat transfer
coefficient, the accurate measurement of temperature was a critical factor in the
experimental setup.
After several runs to test the capacity of the evaporator system, ranges for the selected
variables were chosen. These ranges are listed in Table 4. 1 .
Table 4.1. Ranges of experimental variables
Variable Range
Flow rate (m3/s) 6.67x1O·6 - 6.67xlO -5

Evaporating temperature (OC) 40 - 70
Rotating speed (rad/s) 0 - 21 0
Temperature difference (K) 5 - 50
Feed liquid (No. 1 ) Water
(No.2) Sugar solution (20 - 60%)
(No.3) Reconstituted skim milk ( 1 1 %)
The ranges were suitable for all three evaporation systems utilised (Centritherm, rotating
cone and rotating tube). To avoid excessive concentrate entrainment in the vapour in
trials using skim milk, high temperature differences (greater than 30°C) were not used.
4.7 Operating procedure of thin film evaporator equipment
4.7.1 Starting
( 1 ) Close all valves and set the feed tube at correct position;
(2) Turn on water to the condenser and the vacuum unit;
(3) Turn on the temperature recorder;
(4) Turn on the motor, which makes surface rotate;
(5) Start the vacuum pump;
(6) Check sealing of steam j acket and evaporating side (the pressure should be below
-95 kPa);
(7) Start feed pump and check that the product enters the evaporator, adjust the feed
flow rate to the desired value;
(8) Start concentrate pump and condensate pump, adjust the temperature of evaporating
side through vacuum;
(9) Open drain valve and drain condensate from the stearn supply line, close the drain
valve;
( 1 0) Open stearn valve slowly, adjust steam pressure with the regulator until correct
temperature is obtained;
( 1 1 ) Open cold water valve to water mixer and then open the steam valve to it, adjust
the hot water temperature to the desired value, which is the feed temperature just befo re
entering the feed tube.
4.7.2 Cleaning
( 1 ) Release vacuum in the evaporating side;

(2) Rinse the equipment with tap water;
(3) Recirculate 1 % acid solution for 10 minutes;
(4) Rinse with tap water;

(5) Recirculate 1 % alkali solution for 10 minutes;
(6) Rinse with tap water;
(7) If necessary, recirculate 1 .5 % sanitiser for 1 0 minutes.
4.7.3 Stopping
( 1 ) Set the steam regulator to its minimum value;

(2) Close valves on the steam line and cold water line to the hot water mixer;
(3) Close steam supply valve;
(4) Release vacuum in the evaporating side;
(5) Stop vacuum pump and close supply vacuum valve;
(6) Stop feed and concentrate pumps;
(7) Stop condensate pump;
(8) Switched off motor;
(9) Close valve for cooling water to condenser and vacuum pump;
( 1 0) Switch off the temperature recorder.
4.7.4 Operating cautions
( 1 ) Never run the feed pump or the concentrate pump without liquid. If allowed to run
dry for few seconds, the rubber stator in the pump will be destroyed;
(2) Maximum steam pressure before the steam regulating valve is 600 kPa,
maximum steam pressure after the steam regulating valve is 200 kPa, corresponding to
120°C;
(3) The condenser should be supplied with enough cooling water to keep the outgoing
water temperature 1 0°C below the evaporation temperature;
(4) Always keep an eye on temperature recorder during the experiment, because
adjustment may be needed when desired temperature shifts due to fluctuation in steam
and cold water pressures.
(5) Super heat was eliminated by cooling the outside of the steam line when super heat
was present.
4.8 Experimental Procedure
In order to find the effect of individual variables on the heat transfer coefficients, only
one parameter was varied during each run, while other parameters were kept constant.
A typical procedure was as follows:

( 1 ) Start evaporator with water, and set steam temperature, evaporating temperature, feed
flow rate, feed temperature and rotating speed at the desired values.
(2) After steady state was reached (about ten minutes), the feed was then switched to
the solution to be used. The necessary adjustments were made to maintain the variables
at desired values.
(3) The flow rate of vapour condensate was measured. Flow rates were calculated by
using the values collected during a two minutes period at five minutes intervals during
each test run. That means that the flow of condensate was diverted to CbIempty bucket
and at the same time, time was recorded. After two minutes, the amount of condensate
in the bucket was measured. For each run, three sets of measurements were usually
made. If the error between the each measurement was greater than about 5%, further
measurements were made.
(4) During the experiment, all temperatures were recorded on the temperature recorder.
All data was recorded on a log sheet.
(5) After a test was completed, the feed was switched to water to flush out the
remaining solution in the system, and then cleaning was done.
Initially, the feed temperature to evaporator was very difficult to control due to
fluctuation of the pressure of cold water and steam in the water mixer (preheater). It
became even more difficult to control when the feed temperature was very low before
entering the preheater. Two pressure gauges and two pressure reducing valves were
then installed on the cold water and steam lines respectively to monitor the feed
temperature. The concentrate and the vapour condensate were fed back to feed tank
after their weights were measured. This procedure kept the concentration of feed
constant when running sugar solutions and reconstituted skim milk.
The feed temperature was kept ± 1 °C around evaporating temperature during each run.
For falling film evaporator, the temperature difference between probes No.8 and No.9
was kept less than O.SOc. When the temperature difference was greater than O.5°C,
which indicated that the concentrate might be accumulated at the bottom of the vapour
chamber, adjustments were made.
A mass balance of the feed, concentrate and condensate was carried out at beginning of
each run to check proper vapour condensation and accurate measurement.
4.9 Data processing
Based on the measured feed rate and vapour condensate flow rate, irrigation density as
well as Reynolds number at the top and bottom of the heating tube were computed
according to Equation 3- 1 .
The viscosity of sugar solution was computed according to the equation proposed by
Campanella ( 199 1).
The overall heat transfer coefficient (Hexp) was computed using vapour condensate flow
rate and overall temperature difference (i.e. the temperature difference between probes
No.3 and No.8 in the falling film evaporator and between No.2 and No.3 in the
Centrithem evaporator) measurements. Hexp was calculated from the following formula:
(4- 1 )
Where: Wexp is vapour condensate flow rate (kg/s);

ilrg is the latent heat of vapour condensation, which is a function of the
evaporating temperature (kJ/kg).

A Quattro Pro program, supplied by Honghton Mifflin Company ( 1 993), was used for
all data processing.
The raw experimental data and the results of calculations are given in Appendix I.
CHAPTER 5
DEVELOPMENT OF A THEORETICAL MODEL TO
DETERMINE HEAT TRANSFER COEFFICIENTS
ON ROTATING CONE SURFACES
5.1 Introduction
To gain a better understanding of the heat transfer through a liquid film on a rotating
conical surface, it is necessary to develop a theoretical model for calculating the local
liquid film thickness and liquid side heat transfer coefficient, as well as the overall heat
transfer coefficient. In a rotating surface evaporator, the heating medium is usually
steam, which is condensed on the outside surface of the rotating cone. The liquid to be
concentrated is fed onto the inner surface. Steam condensation on the rotating surface
has been studied intensively by Vasiliev and Khrolenok (1993). However, there are very
few publications in the literature apart from Nikolaev et al. ( 1 967) and Martynov ( 1984)
dealing with the theoretical analysis of the liquid film flow on the inner surface of a
rotating cone. Recently, Bouman and Waalewijn ( 1994a, b) presented a relationship of
the liquid film thickness, the Reynolds number and the centrifugal acceleration on the
rotating cone. The correlation was expressed as a Nusselt number function of the
Reynolds number for concentrated whole milk.
5.2 Theory
5.2.1 Liquid fIlm flow and energy analysis
A schematic of the system used in this study is shown in Figure 5 . 1 .
As shown in the figure, a cone rotates about its vertical axis and the liquid is fed onto
the inner surface. The co-ordinates used in the analysis are: x measures the distance
Chapter 5 The development of a Theoretical Model 68
. . . . .. .. .
.. ... .�<p
. . .. .. . .. . . . ..
.
�
: Q
Figure 5.1 Schematic diagram of the system used
in the theoretical analysis

Chapter 5 The Development of a Theoretical Model 69
along the cone surface from the apex; y measures the distance nonnal to the inner
surface; and cp gives the angular position. The top angle of the cone is 2�. Under the
action of gravity and centrifugal field, the liquid moves along the conical surface, in the
x-direction, as a film.
The boundary-layer model for the conservation of mass, momentum and energy in the
liquid film on the rotating cone can be written as (Schlichting, 1 968):
Conservation of mass:
(5-1 )
x-momentum:
aU" au u/ = P -P v g co s �
u + u x _
x ax Y ay x (5-2)
P
cp-momentum:
(5- 3)
Energy:
(5 -4)
Similar equations were used by Chen et al. ( 1990) to solve the film flow of condensate
in a rotating outer conical surface. These equations apply �o axisymmetric conditions,
which means that:
au aucp _
aT °
__ x = 0, _ - 0, = (5-5)
acp acp acp
The y-momentum equation has not been included here since it only determines the
pressure distribution along the film, and does not have influence on the heat-transfer
calculation. To complete the statement of the problem, boundary conditions have to be
imposed. They are:
ux
= 0'
Uy = 0 ,
at y = 0, Ucp = r n = ( x sin B ) n, (5-6a)
T = Tws
au
x = 0, 't =0
yx
__
ay
at y = 8, aucp _
- 0, 't =0 (5-6b)
yep
-
ay
T T =
evp
The boundary conditions at y = 0 assume that there is no liquid slip at the surface and
the temperature of the liquid film is equal to the wall temperature. At the free surface
of the liquid film y = 8, the boundary conditions assumption is that the effect of shear
by the vapour inside the cone on the liquid film is negligible and the temperature of the
film surface is equal to the evaporation temperature. Sparrow and Gregg (1 959)
demonstrated that for large Prandtl numbers these boundary . conditions gave a good
representation of the system. They also showed that for Prandtl numbers greater than
one, a condition meet by most of the practical situations, energy transfer by convection
is negligible.
Equations (5-2), (5-3) and (5-4) can be further simplified after deleting energy
convection and assuming that:
U2
<9 au au
_ » u x __
+ u __
x (5-7)
x x ax Y ay
Equation (5-7) can be obtained from the continuity equation (Eq (5- 1 » and assuming
that the secondary flow given by the velocity Uy is smaller than U2<9 . To simplify
equation (5-3), the boundary layer theory which considers that viscous effects are larger
than fluid inertia inside the liquid film is assumed. Therefore, Equations (5- 1 ) through
(5-4) become:
(Pux p -p U<92 (5-8)

v __ + __
v g cos � + _ = 0
ay 2 p x
(5-9)
(5-1 0)
The boundary conditions become:
at y = 0, { � : �;
T = T
ws
n = (x sin � ) n , } (5- 1 1 a)
au
x = 0,
ay
at y = 0, aU<9 _
_ - 0, (5-1 1b)
ay
T = Tevp
------ - ------
The solutions of Equations (5-9) and (5 - 10) associated with the boundary conditions
(Eqs (5- 1 1 a,b)) are:
(5- 12)
Tws -Tevp y
T = Tws -
o (5-1 3)
Substitution of Equation (5-1 2) into Equation (5-8) yields the velocity distribution in the
x direction, Ux:
u
x
=
(p - P v g cos �
P v
(5- 14)
To obtain a dimensionless equation, dimensionless variables are introduced.

Dimensionless distances, x· and y . , are as follows:
x y.
x = _, = (5-1 5)
�
•
L L
Whereas dimensionless velocity (Ux·) and dimensionless film thickness (0") are defined
as follows:
o
o· (5-1 6)
L
The characteristic distance (L) and characteristic velocity (Urn) are defined as:
L =
( ) 2
V
1
,- , Um = � (5- 17)
g COSP L
By substituting, Equation (5- 14) becomes:
(5- 1 8)
Ro represents the ratio between the centrifugal and the gravity forces, and it is defined
as:
=
L ( n sin P )2
Ro (5- 1 9)
g cos p
The liquid feed flow rate (Qf) can be calculated as:
B �
Qf = J 2 1t X Sin p Ux dy = 2 1t L 2 Um J Ux' x ' sin p dy ' (5-20)

o 0
Substitution of Equation (5- 1 8) into Equation (5-20) yields the dimensionless film
thickness:
1
,.
o' (x ') =
2 1t L 2 Um x ' sin P
p - P
v + R0 x
(5-2 1 )
By substituting Equations (5- 1 5) , (5- 16) , (5- 1 7) , (5- 1 8) and (5- 1 9) into equation (5-20)
results in:
1
"!
B (x) =
2 1t g x sin � cos �
P - Pv (5-22)
P
5.2.2 Determination of the local heat transfer coefficient
The local heat flux to the liquid film can be calculated using the Fourier's Law:
q - (k �T J
=
y y=o
(5-23)
and the local liquid film heat transfer coefficient is defined as :
hi =
-=-T --q-=-Tevp-
wi
(5-24)
Since the temperature profile across the liquid film is linear (Eq.5- 1 3), the local heat
transfer coefficient can be obtained from Equations (5-23) and (5-24) as:
hl (x) = -
k
B (x)
(5-25)
Substituting the film thickness given by equation (5-22) into equation (5-25) and
integrating from the cone apex (x=L1) to an arbitrary value of x, the local heat transfer
coefficient can be calculated as:
5.2.3 Effect of liquid evaporation
During the evaporation process, the liquid film flow rate on the surface of the cone is
reduced by the liquid vaporisation, which has to be taken into account. The evaporation
rate from the inner surface of the cone can be written as follow:
I I
Weal (X) = h Jhi 2 1t X sin � (Twi - Tevp ) dx ( 5 -27)

Ig L,
'
and the corrected local heat transfer coefficients (hl (x)) is calculated as :
5.2.4 Overall heat transfer coefficient calculation
The overall heat transfer coefficients (Heal) can be calculated considering the resistance
to the heat flow of the steam condensate, the liquid film and the wall of the cone
(thickness ()w and thermal conductivity kv):
----.- - --------
(5-29)
Where: hs is the steam side heat transfer coefficient, which can be calculated using the
Equation (5-30) proposed by Chen et al. ( 1 991).
1
hs !... ( �
=
Pr R )4
L 3 Ja 0 (5-30)
Pr is the Prandtl number defined as:
Pr = _ C
P_
k
Il (5 -3 1 )
Ja is the Jacob number defined as:
(5-32 )
h'fg is the effective latent heat defined as:
h� hfg
� ( 1 + 0.68 (5-33)
Equation (5-30) is suitable for the rotating-dominated film condensation, which is

applicable to thin film evaporators with rotating surfaces.
Calculations of surface areas on the rotating cone are given in Appendix II.
5.2.5 The relationship between rotational Nusselt number and Reynolds number
If p » Pv, then p-pjp= l , Equation (5-22) could be rewritten as:
( 1
J
3 Qfv 1"
=
(5-34)
o
2 rc r g ' + 2 rc r a
Where: r = x.sin�; g'= g.cos�; a= .Q2.r.sin�
If the Reynolds number is defined as Re=4r1/l, where r=QrP/2rcr, the equation can be
rearranged as:
( 1
J
3 v2 �
(11J
=
1" 4r (5-35)
o
4 (g ' + a)
By substituting the Equation (5-25) into Equation (5-34), the following relationship can
be derived:
( 1
J
k 3v2 1"
= Re � (5-36)
hI 4 ( g ' + a)
Therefore the relationship between a new defined rotational Nusselt number (Nuro) and
the Reynolds number can be expressed as:
_
h
( 1
1
J
1" _
=
v2 =
(5-37)
Nu I 1 . 1 0 Re 1"
'0 k g' +a
The above equation is a basic equation that can be used to predict local film heat
transfer coefficients for rotating conical surface under laminar film flow and surface
evaporation.
The liquid film Nusselt number (Nu) can be calculated using the dimensionless film
thickness as follows:
h/ L L 1
(5-38)
Nu = = =
k "8 a-
5.3 Calculation procedure
To solve the equations given in the previous section, a numerical integration procedure
was implemented in an applied mathematics software package (Mathcad 3 .0, 1 993). An
iteration was started by assuming a initial value of wall temperature on the liquid side
(Tw I) as the mean value of the steam condensing and the liquid evaporating temperatures,
i.e. Twl=(Ts + Tevp)/2, and solving for the wall temperature on the steam side (Tws), steam
side heat transfer coefficient (hs), liquid side heat transfer coefficient (hI) as well as the
overall heat transfer coefficient (Heal) until convergence was finally obtained. A flow
diagram for calculation is shown in Figure 5.2.
5.4 The determination of the physical properties of the liquids used
Three liquids were used in the experimental work: water, sugar solutions and skim milk.
The physical properties of water are only a function of temperature and the physical
properties of sugar solutions and reconstituted skim milk are mainly functions of the
temperature and their concentrations. The correlations used for the theoretical
calculation are given in the Appendix III .
Chapter 5 The development of a Theoretical Model 79
Select the values for
Tovp. Ts. B . Xl. X2
Cal culate W eA l
using Eq (5-27)
Calculate h'l
using Eq (5-28)
C a l cu l a te h.
using E q (5-30)
R e c alculate T wI2
No
� Yes
C al c u l a t e Hell using Eq (5-29)
Figure 5.2 Flow diagram for the numerical calculation of the
overall heat transfer coefficient

Heat transfer and Fouling in Film Evaporators with Rotating Surfaces 80
CHAPTER 6
HEAT TRANSFER IN FILM EVAPORATORS
RESULTS AND DISCUSSION
6.1 Theoretical analysis results
The theoretical Equations (5-22) and (5-29), to predict the liquid film thickness and the
heat transfer coefficient derived in the Chapter 5, show that the variables affecting the
liquid film thickness and the overall heat transfer coefficient are: the feed flow rate, the
cone rotating speed, the length of the cone, the cone angle, and the liquid viscosity.
To reveal the effect of these variables more clearly, the results obtained are presented
graphically. The conditions used in the numerical calculations are those which are
relevant to the Centritherm evaporator, i.e. rotating speed 1 57 rad/s, feed flow 5 x 1 0-5
m 3/s and evaporating temperature 70°C. Water was considered as the evaporating liquid
for this part of the calculation. The results of the numerical calculations are given in
the Appendix IV.
Figure 6. 1 shows the local dimensionless film thickness for different cone angles. As
the distance along the cone increases, the film thickness decreases due to the increase
of the surface area, centrifugal acceleration and vaporisation. This figure also shows
that the film thickness decreases with increasing cone angles. This is obviously a result
of the increasing acceleration component in the direction along the cone surface.
Figure 6.2 shows the effect of feed flow rate on liquid film thickness. As the feed flow
rate increases, the film thickness also increases_ As depicted in the figure, this effect
CHAPTER 6 Heat Transfer in Evaporators--Results and Discussion 81
50
• C o n e a n g l e f3 =
1 0·
• A C o n e angle f3 =
20·
c.o
40 =
T C o n e angle f3 30·
II)
=
* C o n e angle f3 40·
II)
Q)
c:
=
u
.x
30 • C o n e angle f3 50·
:E
-
E
- 20
II)
II)
Q)
c:
0
II)
10
c:
Q)
E
CS 0
1 00 1 000 1 00 0 0
Dimensionless distance a l o n g t h e cone ( X· )
Figure 6_1 Dimensionless f ilm thickness o n a r otating c o n e a t
different cone angles_ R o t a t i n g s p e ed: 1 5 7 r a d / s ,
f e e d flow: 5 x 1 0- s m3 /s, w a t e r a t 70 · C.
25
• feed flow: 6 . 7 x 1 0- 5 m'/s
•
&0
20 • feed flow: 5.0 x 1 0- 5 m'/s
'"
'"
T feed flow: 3.3 x 1 0- 5 m'/s
A x 1 0- 5 m'/s
Q)
c feed flow: 1 .7
�
0 15
-
.£:
E
- 10
'"
'"
Q)
c
0
'"
5
c
Q)
E
0
0
1 00 1 000 1 0000
Dimensionless distance along t h e c o n e ( X· )
Figure 6 .2 Dimen s i o n l e s s film thickness o n a r o t a ting cone a t
different f e e d flows. Rotating s p eed: 1 57 rad/s,
cone angle: 40·, water at 7 0 ·C .

is more pronounced at the apex of the cone (lower values of x· ), whereas results appear
to approximate to the same value towards the exit of the evaporator (large values of x* ).
Figure 6.3 shows the effect of rotation speed (defined by the rotation parameter Ro) on
the dimensionless film thickness at a specific position (the middle of the conical
surface). It clearly shows that the film thickness decreases with the increase of rotating
speed. The figure also illustrates that this effect is reduced for higher cone angles.
The decreasing rate of dimensionless film thickness is bigger when rotation parameters
are less than 0.01 than when rotation parameters are greater than 0.0 1 . This implies that
the effect of the gravity force becomes less important when the rotation parameter is
bigger than 0.0 1 .
Figure 6.4 shows the effect of rotating speed on the Nusselt number as a function of the
cone angle . The Nusselt number increases as the rotating speed and cone angle
increases.
The calculated local heat transfer coefficients as a function of the position for each of
the heat resistances on the rotating cone (using conditions for the Centritherm
evaporator) are shown in Figure 6.5. Conditions used in the calculation are rotating
speed 1 57 rad/s, feed flow rate 5xlO-5 m3/s, evaporating temperature 70°C and
temperature difference 10K. It can be seen that in the liquid side the heat transfer
coefficient can vary from 1 2 to 23 kW/m2K, which is due to the increase in the surface
area and the evaporation of the liquid. W"hile the steam side heat transfer coefficients
are in the range 55-60 kW/m2K. As a result the overall heat transfer coefficients
increase slightly along the cone surface from the top to the bottom.
Figure 6.6 shows the calculated local surface temperature of the cone on both sides, the
local liquid film thickness and the Reynolds number on the inner surface of the rotating
cone with the same conditions as that used in the Figure 6.5. It can be seen that when
steam temperature is 80°C and evaporating temperature is 70°C, the surface temperature
1 5.0
1 2.5 • Cone angle B = 1 0°

A Cone angle B =
2 0°
• ... Cone angle B =
30°
ro
1 0.0
• Cone angle B =
4 0°
* 5 0°
en
en Cone angle B =
Q)
c:
.:.t:. 7.5
0
-
.r:::.
E
- 5.0
en
en
Q)
c:
0
en
2.5
c:
Q)
E
is
0 .0
1 0- 5 1 0-4 1 0- 3 1 0-2 1 0- 1 1 0° 1 0'
Rotation p a r a meter ( Ro )
Figure 6.3 Dime nsionless film thickness versus r o t a ti o n parameter
a t different cone a n g l e s . Feed flow: 5 x 1 0- 5 m 3 /s,
dimensionless distance (X+): 5.57x 1 03 , w a t e r a t 7 0 · C.

1 .75 * Cone angle B = 50·
• Cone angle B = 40·

1 .50 *
T Cone angle B = 30·
... Cone angle B = 20·

1 .25
• Cone angle B
•
GO 1 .00
......
.-
�
Q)
.0
0.75
E
;:,
c:
0.50
Q)
'"
'"
;:,
z 0.25
0.00
o 50 1 00 1 50 200 250 300 350
Cone rotating speed (rad/s)
Figure 6.4 Effect o f rotating speed on the Nusselt number a t
different c one angles. F e e d f l o w : 5x 1 0-5 ml/ s , water
at 7 0 · C, p osition given by X·=5.5 3 x 1 03 (X=0. 1 55m).

70
� + ----
C\I 60
E +-- S t e a m s i d e hs
'- +-- -
3= + -+
..::t:.
-
c 50
Q)
()
-
-
Q)
0
()
�
40
Q)
-
(I)
c
ro
�
-
- 30
Cone (kw / Sw)
ro
Q) T T
:r:
. ...
. ' . . .
. . . ... .
.
.. . .
20 l i q u i d s i d e hi
. ' .
.... . . .
. .
. .
... . .
• •
10 •
•
•
O v er a l l h e a t t r a n s f e r
c o e f fi c i e n t (Heal)
0.05 0.1 0 0.1 5 0.20 0 .25
D i s t a n c e along the c o n e ( m
Figure 6.5 C a l c ula ted h e a t t r a n s f e r c o e f f i c i e n t s on the r o t a t i n g c o n e
i n t h e C e n t r i t h e r m e v a p o r a t o r . R o t a t ing speed: 157 rad/s,
f e e d flow: 5 x 1 0- 5 m3 / s, e v a p o r a ti n g t e m p e r a ture: 7 0 oe,
temperature difference: 1 0K, e v a p o r a t ing liquid: w a te r .

CHAPTER 6 Heat Transfer in Evaporators--Results and Discussion
87
80
. ------ . ------ ------ ------

a> . . •
L- S t e a m side
::l
-
IV
L-
a>
0. _ T ___
E () T ___
� :.... 75 ��
T-
-T -
-__-
a> a> T Liquid side
o c:
IV 0
L-
_ 0
::l _
C/) 0
70
- 6
E
\I)
I
0
......
-)( A �
E 4
-
en
en A _____
A
a>
. .:.:.
"0 c:
-
::l ____ A
.-
0
CT .-
.&:
___ A
...J -
2
L- 350
a>
.0 300
E
::l 250
c:
en 200
"0
'0 1 50
c:
a>
>-
1 00
c:
0.05 0. 1 0 0. 1 5 020 0.25
Distance along the cone ( m )
Figure 6 . 6 Ca lculated local surface tempe ratures, liquid film thickn e s s and
Reynolds number along the cone. C onditions used in t h e calcula tions
w ere those of the Centritherm evaporator. Rotating s p eed: 1 86
rad/s, evaporating temperature: 70·C, tempera ture diffe r ence:
1 0K, feed flow: 5 x 1 0-5 m3/s, evaporating liquid: water.

varies between 78.7 to 75. 1°C on the steam side and between 75.8 to 74.4°C on the
liquid side. The liquid fIlm thickness is in the range of 0.025-0.044 mm . These results
are in the same order of magnitude to the values (0.04-0.08 mm) reported by Hallstrom
( 1 977) for water. Reynolds numbers for the inner surface liquid fIlm are in the range
of 150 to 350.
6.2 Experimental results and their comparison with those theoretically calculated
with the model
Measured overall heat transfer coefficients were calculated as indicated in Chapter 4

(section 4.9). The uncertainty of the overall heat transfer coefficients calculated from
these measured variables was detennined to be 0.55 kW/m2K. Therefore, the 95%
confident interval of the measured heat transfer coefficients is around ± 1 0% (an example
of calculations is given in the Appendix V). By this reason, the error bar was not
included in the figures.
Figure 6.7 shows typical recorded curves of steam temperature, evaporating temperature
and feed temperature during an experimental run, which indicates the accuracy of the
temperatures monitored.
An attempt was made to measure the surface temperature of the rotating surface by
using a thermocouple, but it was not successful due to the heat generated by friction
between the thermocouple and the rotating surface. Therefore, it was not practical in
the setup used to measure the surface temperature of the rotating surface, and only
overall heat transfer coefficients were calculated. The condensing side heat transfer
coefficients were calculated using equation (5-28) and the values were in the order of
60 kW/m2 K under the conditions of this experiment (as shown in the Figure 6.5). These
coefficients are considerable higher than the fIlm coefficients on the evaporating side.
Although the heat transfer resistance of the cone itself was calculated to be in the
1 20
1 00 -
-
U
- - .... ... � ... ...
•
80 - - - -
- -
Q)
�
-
:::l
� A
IU
..... -- A
�
�
Q) --.;;J'"
a.
E
Q)
t-
60 �
40 1 I I I I
30 45 60 75 90 1 05 1 20
Time (minutes)
Figure 6 .7 Recorded temperatures during an experimental run in
the Centritherm evaporator.
C Steam condensing temperature
h. Liquid evaporat i n g t empera ture
o Feed temperature
similar order of magnitude as the liquid film resistance, the evaporating liquid film is
still the major controlling factor during the heat transfer process.
Experimental results obtained with the Centritherm, the cone evaporator and the single
tube falling film evaporator with rotating surface are presented in the following sub
sections.
6.2.1 Centritherm evaporator
Experiments carried out using the Centritherm evaporator with water, a 20% sugar
solution and skim milk are presented in this section.
A 42 factorial experiment with water was first conducted to get a basic understanding
of the effects of selected variables, namely temperature difference between the steam
condensing temperature and the liquid evaporating temperature, rotating speed, feed flow
rate and liquid evaporating temperature. The following correlation was obtained using
a multivariate analysis of the experimental data in the spreadsheet Quattro Pro:
Hexp = 0.289 + 0.055�T + 0 1 8 Q

.
+ 0. 1 Tevp + 0.055 Qf (6- 1 )
(R2 = 86.4%, df=49)

where: �T is the temperature difference between the steam condensing
temperature and the liquid evaporating temperature (K)
.Q is the rotating speed of the cone (rad/s)
Tevp is the evaporating temperature (OC)
Qf is the feed flow rate (m3/s)
(The t values for each variables are trevp= 1 5 ; tt.r=7.75; to=5. 15; tQFO.025)
The results obtained show that the effects of temperature difference, evaporating
temperature and rotating speed on the overall heat transfer coefficient were significant,
while the feed flow was less significant. The effect of each of these variables was
investigated independently with the different liquids listed above.
6.2.1.1 Temperature difference
Figure 6.8 shows curves of measured overall heat transfer coefficient against the
temperature difference between the steam condensing and the liquid evaporating
temperatures for water, a 20% sugar solution and skim milk. Results obtained from the
theoretical model were also plotted in the same graph. Solid lines indicate measured
overall heat transfer coefficients, whereas dashed lines represent the theoretically
calculated values, using Equation (5-29) under the same conditions of the experiment.
It can be observed that both measured and theoretical calculated values are close and the
trend between measured curves is similar to that of the theoretical curves up to
temperature differences of about 30K.
The dashed lines show that the calculated overall heat transfer coefficients increase
slightly with the increase of the temperature difference. As the evaporation temperature
was kept constant, an increase of temperature difference implies that the steam
temperature and consequently the wall temperature of rotating cone is also increased.
This affects the physical properties of the evaporating liquid and results in a slight
increase of the overall heat transfer coefficient.
The solid lines show that the measured heat transfer coefficients increase with the
increase of temperature difference until 30K and then decrease. This result suggests that
an evaporation liquid-vapour interface may occur only at low temperature differences,
and under that regime the heat transfer coefficient increases as the temperature
difference increases. At constant feed flow, the increase in temperature difference
causes more vapour to be produced, which in turn results in a reduction of the liquid
film thickness and consequently an increase in coefficient of heat transfer.
� 10
N
E
'"
3:
..lII:
- 9 .... - -
c .-
Q)
0
-
-
Q)
0
0 8
....
Q)
-
en
"'
C
.... • Water
-
-
7
"' • Sugar solution ( 2 0 % )
Q)
.c:.
T Skim milk ( 1 1 %)
-;;
....
Q)
> S
0
0 10 20 30 40 50
Temperature difference K
Figure S .8 Effect o f tempera ture difference on the overall h e a t
transfer coe fficient f o r water, 2 0 % s u g a r s ol u tion a n d
skimmilk in t h e Centritherm evapo r a t o r. S olid lines
indicate experimental values while d a s h e d lines indic a t e
theoretical values. E v apora ting t e m p e ra ture: S O · C , f e e d
flow: 5x 1 0-5 m3 /s, r o tating speed: 1 8 6 rad/s.

CHAPTER 6 H eat Transfer in Evaporators--Results and Discussion 93
Even though experimental and theoretical values appear to agree in some extent there
is a degree of discrepancy between them. This discrepancy may be attributed to bubble
fonnation in the evaporating liquid film. At low temperature differences, there may be
incipient bubble fonnation and collapse. When bubble fonnation increases as a result
of increasing temperature difference, bubbles tend to drag liquid with them and hence
provide convection currents assisting the transfer of heat to the surface. However, the
presence of bubbles could thicken the film and reduce heat transfer to the liquid-vapour
interface. These two opposite effects result in that the heat transfer coefficients
increases with the temperature difference between steam and liquid, but the rate of
increase is reduced at higher temperature differences.
For water and sugar solution, it can be seen that the heat transfer coefficient decreases
with further increase in temperature differences after 30K. This is probably because at
high temperature differences, bubbles fonnation is very high and bubbles may start
sticking to the surface and reducing the available surface. Hence the heat transfer
decreases. Skim milk was not used when temperature differences were over 30K due
to fouling of the surface.
6.2.1.2 Rotating speed
Figure 6.9 shows calculated (dashed lines) and experimental (solids lines) overall heat
transfer coefficients as a function of the rotating speed in the Centrithenn evaporator for
water, a 20% sugar solution and skim milk. The increase of heat transfer coefficient
with increases in the rotating speed is depicted in this figure. This indicates that the
component of centrifuge force along the rotating surface plays a very important role in
the heat transfer. As the rotating speed of the rotating surface increases, the centrifugal
force is increased, then the speed of the liquid moving in the direction parallel to the
surface will increase, and the film thickness will reduce, which results in a lower
resistance to the heat transfer.
CHAPTER 6 Heat Transfer in Evaporators--Results and Discussion
94
- 1 0.0
�
N
E
......
3:
.:.:. 8.7
-
t::
a)
o
-
_ 7.4
a)
o
o
L-
a)
"en 6. 1
t::
I'G
L-
-
-
• Water
I'G
a) 4.8
.t:: ... Sugar s olution (20%)
I'G
L-
a) � Skimmilk ( 1 1 %)
>
o 3.5
o 50 1 00 1 50 200
R otating speed ( rad/s
Figure 6.9 Effect of the cone rotating speed o n the overall heat
transfer c oeficient for w at e r, 20% sugar ' solution and
s kimmilk in the Centriterm evaporator. Solid lines are
experimental values while dashed lines a r e theoretical
v alues. Evaporating tempe r ature: 70 ·C, feed flow:
5x 1 0- 5 m3/ s, temperature difference: 1 0K .

Rotating the cone also beneficially influences heat transfer on the steam side. As stearn
is the heating medium, it condenses in the form of droplets (Mannheim and Passy, 1974)
and is impelled from the surface. It can be deduced that the diameter of condensate
droplets impelled from the rotating surface decrease as the speed of rotation rises. So
the resistance to heat transfer lowers when the rotating speed is increased. It can be
seen that both calculated and experimental overall heat transfer coefficients follow the
same trend when the temperature difference is 1 0K. This means that for this
temperature difference the heat transfer through the thin liquid film on the rotating
surface is dominated mainly by the conduction and the main mechanism of evaporation
is from the surface of the liquid film. If nucleate boiling contributed significantly to the
mechanism of evaporation, there would be a change in the slope of curves in Figure 6.9,
such as occurs in falling film evaporation (Chen and Jebson, 1 992). MaIkki and
Veldstra ( 1967) also stated that evaporation in the Centritherm evaporator takes place
mainly from the liquid film surface.
Figure 6.9 shows there are some discrepancies between theoretical and experimental
values at low rotating speeds. At low rotating speeds the film is comparatively thick.
This encourages wavy laminar flow which increases the heat transfer. It was observed
by Shinn (197 1 ) that waves exist in the liquid film when it moves across the rotating
conical surface. But when the speed increases the film thickness reduces the wave
effect bringing the heat transfer coefficient closer to the theoretical value as observed
in Figure 6.9.
6.2.1 .3 Feed flow
Figure 6. 10 shows the effect of feed flow rate on both experimental (solids line) and
theoretical (dashed lines) overall heat transfer coefficients for water, 20% sugar solution
and skim milk. Experimental results show a slightly increase in heat transfer coefficient
with increasing feed flow rate. Two conflicting effects may be occurring in this
experimental range. As the feed flow rate increases, the Reynolds number also
C HAPTER 6 Heat Transfer in Evaporators--Results and Discussion 96
12
�
N
E • Water
.......
3: 11 ... Sugar solution (20%)

�
-
c:
" T Skimmilk ( 1 1 % )
Q)
10
'(3
-
- �',
----
�,
Q)
�, • •
. ''-
" :::e., � ""
0
0 9
L-
-
Q)
.....
CI)
c:
cv
L-
-
-
8
� ..t >F "' '':
"" ",,- --�
cv
Q) � "" ......
.r:
7 -- �
cv
L-
Q)
>
0 6
0.0 1 .5 3.0 4.5 6.0 7.5
F e e d f l o w rate x 1 0-s m 3 / s
F i g u r e 6. 1 0 E f f e c t o f t h e f eed f l o w r a t e on t h e overall h e a t t r a n s fer
coefficient for water . 20% sugar solution and skimmilk
in the C entritherm evapora tor. Solid lines are
experi m e n t a l v a l ues while dashed lines are the o r e ti c a l
values. E v a p or a t ing temperature: 6 0 · C, rotating
speed: 1 4 6.6 rad/s, tempera ture d i ff erence: 1 0K.

increases, which would increase the wave motion and turbulence, hence, increasing the
heat transfer coefficient. However, the increase of feed flow also increases the film
thickness, decreasing the effect of liquid-vapour interface evaporation and lowering the
heat transfer coefficient. The theoretical calculated values, which are based on the assumption of
a laminar flow regime, illustrate the latter effect . Thus, the theoretical lines show a decrease
in heat transfer with increasing flow rate. The combination of the above results shows
that the heat transfer coefficient should increase slightly with increasing of feed flow.
This result is in agreement with those obtained by Yanniotis and Kolokotsa ( 1 996).
They found that the feed flow rate had no significant influence on the heat transfer
coefficient obtained from the heated surface of a rotating disc (feed flow in the range
of 1 .67 x 10-5 - 8.33 X 10-5 m3/s, rotating speed in range of 2 1 - 63 radls) .
Figure 6. 1 1 shows the effect o f the feed flow and the rotating speed on the heat transfer
coefficient . The interaction between feed flow and rotating speed on the heat transfer
coefficient was not significant under these conditions. By using the data of Figure 6. 1 1 ,
a regression equation was obtained:
In ( 9.77 - Hexp ) = 1 .252 - 0.002Q + 0.00 1 Qf - 1 .24 X 10-7 Q/ (6-2)
(R2 = 97.09%, df=9)

The results show that the feed flow has a slight effect of the feed flow on the heat
transfer coefficient. It also appears that it is more important at low rotating speeds and
high feed flow rates.
However, at some extreme conditions, it looks as there is an interaction between the
rotating speed, feed flow and temperature difference. Figure 6. 1 2 shows, for water, that
the heat transfer coefficient increases with decreasing feed flow below 1 .67 x 1 0-5 m3/s
at a temperature difference of 1 0K and a rotating speed of 1 86 radls. The heat transfer
coefficient, however, decreases with decreasing the feed flow below 3.34 x 1 0-5 m3/s for
temperature difference of 30K and a rotating speed of 42 radls. The explanation may
1 0.0
9.5
....
c:
CI)
9.0
(,)
-
-
CI)
o
(,) 8.5
L.
-
CI)
en
c:
C'O
L.
8.0
....
....
C'O
CI)
.c:
7.5
C'O
L.
CI)
>
o 7.0
0 .0 1 .5 3.0 4.5 6 .0 7.5
F e e d flow rate ( x 1 0- s m 3 / s )
Figure 6 . 1 1 E f f e c t of the f e e d flow rate a n d the r o t a tional s p e e d
o n the overall h e a t t r ansfer coe fficient for w a t e r i n
t h e Centri therm evaporator. Evaporating temperature:
7 0 · C . tempera ture dif ference: 1 0K .
T R o tating s p e e d 1 57 rad/s
.. R o t a ting s p e e d 1 05 r ad/s
• R o t a ting speed 52 rad/s

�
N
•
E
..... 1 2.8 "-
3: �
.:.t:
�
....
"
c: "
1 0.6 . ......
Q)
0
;0:: .... ""
-Q)
0
0
� •
•
�
• --:--- - ===-
�
-Q) 8.4
'"
c:
co ... ... ... ...
�
-
*
....
co
Q)
.r;
6.2 � .......
- ""........
co
�
"'"'-"",,-- - 4.-
Q)
> - "-- - -..
0 4.0
0.0 1 .5 3.0 4.5 6.0 7.5
Feed flow r a t e ( x 1 0- s m 3 / s )
Figure 6. 1 2 Effect of the feed flow rate on the overall h e a t transfer
coef ficient for water in the Centritherm e vaporator.
Solids lines are experimental values while d a shed lines
are theore tical values. Evapora ting temperature: 70·C.
• T emperature difference: 1 0K , Rotating s peed: 1 86 rad/ s
... Temperature difference: 30K, rotating s peed: 42 rad/s

be that the reduction in feed flow at a low temperature difference and a high rotating
speed causes a decrease in the fIlm thickness, resulting in the increase of the heat
transfer coefficient. This trend is also shown by the theoretical analysis. However, the
reduction of feed flow at a high temperature difference and a low rotating speed may
cause incomplete coverage of the liquid fIlm on the rotating cone surface, which results
in a decrease of the heat transfer coefficient. For the high temperature difference, the
theoretical model does not predict the trend of the curve shown in Figure 6. 1 2. The
main reason could be attributed to a phenomena of natural convection which increases
the heat transfer coefficient. Bubble formation at high temperature differences and low
feed flow rates may also disturb the film coverage. At high temperature differences
there would be much more bubble formation, to the extent that the measured heat
transfer coefficients are actually higher than the theoretical ones.
6.2.1.4 Evaporating temperature
The effect of the evaporating temperature on the overall heat transfer coefficient for
water, 20% sugar solution and skim milk is shown in Figure 6. 1 3 . It can be seen that
both theoretical and experimental overall heat transfer coefficients increase as the
evaporating temperature increases. At these experimental conditions, both experimental

and theoretical values are within 1 2% agreement. However, the experimental values are
lower than the theoretical values.
The effect of evaporating temperature is attributed to the changes in the liquids ' physical
properties (liquid viscosity and thermal conductivity) with temperature. The values of
viscosity, thermal conductivity and Prandtl number at different temperatures for liquids
studied in this research are given in Tables 6. 1 - 6.3. It can be seen that as the
evaporating temperature increases, the liquid viscosity decreases and the thermal
conductivity slightly increases. The changes in these properties result in a thinner fIlm
with increased heat transfer conduction an therefore a higher heat transfer coefficient.
10
--
.".....-
-
c
--
Q)
--
----
o
-
:;: ----
Q) 8 lit!""' ----
o � -
o
�
-
Q)
o
C
co
-
....
-
co 7 • W a ter
Q)
..r:::.
.. Sugar solution (20%)
co
�
Q)
> y Skim milk ( 1 1 %)
o
50 55 60 65 70 75
(
I
E vaporatmg
· tempera t ure e )
Figure 6. 1 3 Effect of the evaporating t empera ture on the o v e r a l l
heat transfer coefficient in t h e Centritherm
evapora tor. S olid lines are experimental values while
dashed lines are theoretical values. Rotating s p e e d:
1 86 rad/s, feed flow: 5x 1 0- s m 3 /s, temperature
difference: 1 0K.
CHAPTER 6 Heat Transfer in Evaporators--Results and Discussion 1 02
Table 6.1 Physical properties of water as a function of temperature (from Rogers
and Mayhew, 1 984)
Water SO°C 60°C 70°C
Viscosity (Pa.s) 5.4x104 4.6x 104 4.0x 1 04
Thennal conductivity (W/m.K) 0.643 0.653 0.662
Prandtl number 3.9 3 .0 2.5
Table 6.2 Physical properties of a 20 % sugar solution as a function of temperature
(from Campanella, 199 1 )
20% sugar solution SO°C 60°C 70°C
Viscosity (Pa.s) 9.4x 104 7.8x 1 04 6.7xlO-4
Thennal conductivity (W/m.K) 0.642 0.65 1 0.660
Prandtl number 6. 1 5.0 4.2
Table 6.3 Physical properties of skim milk as a function of temperature (from
Wood, 1 982)
Skim milk ( 1 1 %) 50°C 60°C 70°C
Viscosity (Pa.s) 8 .S6x104 7 . 8x 1 04 7.5xlO-4
Thennal conductivity (W/m.K) 0.6 1 6 0.625 0.632
Prandtl number 5.8 5.2 5.0

6.2.1.5 Experiments of sugar solutions at different concentrations
Typical curves of overall heat transfer coefficient versus temperature difference for
different concentrations of sugar solution are shown in Figure 6. 14. Both experimental
and theoretical values are reasonab. ly close except for the highest concentration
(60%TS) at low temperature differences (below 15K).
The evaporation mechanism in this experiment is likely to be a combination of film

surface evaporation and bubble formation on the surface of the rotating cone. When the
temperature difference increases, the heat transfer by conduction will increase, because
thermal conductivity of the film increases slightly. At the same time, more bubbles are
formed and present in the liquid film, which increases the film thickness, decreasing the
rate of heat transfer to the liquid-vapour interface. In addition, the motion of bubbles
through the liquid film encourage the turbulence and has a positive effect on heat
transfer coefficient as mentioned on section 3.2. 1 . 1 . As a consequence of these effects,
the overall heat transfer coefficient slightly increases with an increase of temperature
difference.
The experimental values show that when the concentration rises to 60%, the heat
transfer coefficient reduces significantly with decreasing temperature differences. The

reason for this decreasing is not clear. In general, as the concentration of the solutions
increases, the overall heat transfer coefficient reduce gradually. The heat transfer
coefficients, however, are still around 5 kW/m2K at concentrations of 60% when the
temperature difference is greater than 1 5 K.
Figure 6. 1 5 shows the effect of concentration on the overall heat transfer coefficient
when the evaporating temperature is 70°C, feed flow 5x 1 O-5 m3/s, temperature difference
1 0K and rotating speed 1 86 rad/s. It is clearly illustrated in this figure that when the
concentration increases, the associated increase in film viscosity and thickness decreases
the heat transfer coefficients. Both calculated and experiment overall heat transfer
coefficients follow the same trend and are in reasonable agreement, which again
10
-- - • • . - -
• • •
8 •
-
c::
CP
o
-
0;: 6
CP
o
o
L..
CP
-
en
c:: 4
IV
-
L..
• 30% T S
� 40% TS
2 T 50% T S
IV •
•
L..
CP 60% TS
>
o
o
o 5 10 15 20 25 30
Temperature differ ence ( K )
Figur e 6_ 1 4 E f f e c t o f the tempera ture difference on the overall
h e a t transfer coefficient for sugar solutions i n the
C e n t ritherm evaporator. Solids lines are e x p erimental
v a lues . while dashed lines are theoretical v alues.
R o t ating speed: 1 86 rad/s, feed flow: 4.2x 1 0-s m3 /s,
e v a p orating temperature: 7 0 · C.
10
�
N
E
.......
3= 8
�
.....
c::
CD
0
- 6
-
CD
0
0
....
-
CD
CIJ 4
c::
....
co
.....
.....
co
CD
..r::. 2
-
.... •
co
CD
>
0
0
10 20 30 40 50 60 70 80
C o ncentra tion o f sugar solutions ( % )
Figure 6. 1 5 Effect o f sugar concentration o n the overall h e a t trans fer
coeficient in the Cen tritherm e v a p or a tor. E v a p o ra ting
temp e r a ture: 70 · C, temperature d i fferen ce: 1 0K , f e e d
f l o w : 4 .2x 1 0- 5 m3 /s, rotating s peed: 1 86 r a d / s .
• E x p erimental results (solid line)
T T h e o r e t i c a l resul ts (dashed line)

indicates that the heat transfer through the thin liquid film on the rotating surface at low
temperature differences occurs mainly by conduction and the main mechanism of
evaporation is from the surface of the liquid film.
6.2.2 Cone evaporator
The cone evaporator was specially built to study the effect of the cone angle (formed
by the rotating surface around the vertical axis) on the overall heat transfer coefficient.
Only water was used in the experiments.
Figure 6. 1 6 shows the measured and calculated overall heat transfer coefficients versus
the temperature difference in the cone evaporator. It is found that the temperature
difference, in the range studied, does not affect these overall heat transfer coefficients.
The experimental results follow a similar trend to the theoretical ones, but the heat
transfer coefficients, however, are not in as a close agreement as those for the
Centritherm evaporator.
Figure 6. 1 7 shows the effect of the rotating speed on the overall heat transfer coefficient
as a function of the evaporation temperature. Both the theoretical and experimental
curves follow the same trend. This again indicates that the component of centrifugal
force along the rotating surface plays an important role in the heat transfer. The film
thickness will reduce as the speed of the liquid moving over the surface increases (due
to the increases of rotating speed), which results in a lower resistance to the heat
transfer.
At low rotating speeds the concentrate extraction pump was not able to handle higher
flow rate. Hence it was not possible to make measurements in this region.
10
�
N
E
.......
3:
�
8
- - - - - - •
c: - - - -
CI>
0
-
-
CI> 6
0
0
...
-
CI>
en
c: 4
...
co
- • • •
-
co
CI>
.r:;
2
...
co
CI>
>
0
0
0.0 2.5 5.0 7.5 1 0.0 1 2.5 1 5.0 1 7.5
T empera ture difference ( K )
Figure 6 . 1 6 E ffect o f t h e temperature difference on the overall
heat transfer coe fficient for water in a cone e v a p o ra tor.
The solid line is conne cting ex perimental values a n d
the dashed line i s connecting theoretical values.
Rotating s p e ed: 230 rad/s, feed flow: 4.2x 1 0- 5 m 3 / s ,
evaporating temperature: 70·C.

7.50
� T Evaporating T empera ture: 60 · C
N
E
...... ... Evaporating T emperature: 70·C �
3=
.:Jt:
/ � .,;r
6.2 5 //
- Y"
c: ,J' /
//
a>
(J
;0:: / ;1'
- /
a>
0 / / /t
(J 5.00 //
L-
x/
-
Q)
en / .r
11' /
c:
ft:I
L-
-
/. '"
-
ft:I
a>
3.75 �/
�/
�
.c:
ft:I
L-
a>
>
0 2.50
0 50 1 00 1 50 200 250
R o tating speed ( r a d / s
Figure 6 . 1 7 E ffect o f rotating speed o n the overall h � a t transfer
c oefficient a t different evaporating temperature in a
cone evapor a t or for water. Solid lines are experim e n t al
values while d a s h e d lines are theoretical values.
Tempera ture difference: 1 0K , feed flow: 4.2x 1 0-s m 3 / s .

6.2.2.3 Feed flow
Figure 6. 1 8 shows that the feed flow has only a slight effect on the heat transfer
coefficient in these conditions, the results are similar to those for the Centritherm
evaporator. As the feed flow rate increases, the Reynolds number also increases, which
would increase the wave motion and turbulence, hence, increasing the heat transfer
coefficient. However, the increase of feed flow also increases the fIlm thickness,
decreasing the effect of liquid-vapour interface evaporation and lowering the heat
transfer coefficient. The later effect is illustrated in the theoretical line. The net effect
of these two conflicting effects means that the heat transfer coefficient increases only
slightly with increasing feed flow rate.
Figure 6. 1 9 shows the effect of evaporating temperature on the overall heat transfer
coefficient. It can be seen that both the theoretical and experimental heat transfer
coefficients follow the same trend. The experimental heat transfer coefficients, however,
are lower than the theoretical ones.
The results above reported show that measured overall heat transfer coefficients are
considerably lower than the theoretically calculated values in the cone evaporator. No
explanation for the difference could be found, but it is thought that a poor feed
distribution may be one of the cause for the discrepancies. It is important to note,
however, that in general, the overall heat transfer coefficient experimentally measured
in the cone evaporator were lower than those measured in the Centritherm evaporator,
which indicates that the cone angle has an important role in determining the overall heat
transfer coefficients in this type of evaporator.
6.2.3 Falling fllm evaporator with a rotating tube
The initial idea of rotating the tube in a falling fIlm evaporator arose from the
observation that the heat transfer coefficient in the Centritherm evaporator is much
9
:::c:
N
E 8
�
'- "-
3: ......
.:.: "
7
- ...... -.., --
c .
Q) - - .....
0 6
-
-
Q)
0
0 5
�
-
Q)
CI)
c 4
C'CJ
�
-
- . ------- . ------ .
C'CJ 3
Q)
.s::.
C'CJ
�
2
Q)
>
0
1
0.0 1 .5 3.0 4 .5 6.0 7 .S
Feed flow rate ( x 1 0- 5 m 3 / s )
Figure 6 . 1 8 Effect o f the feed flow r a te o n the o ve rall h e a t transfer
coefficie n t for water in a cone evapora tor. Tempera ture
difference: 1 0 K , rota ting s p e e d : 230 rad/s , e v a porating
temperature: 60·C. The s olid line is connecting
experimental values while the dashed line is c on n ecting
theoretical values.
8
:::.::
N
E
"
�
.:.c
6
.- - - ----
- ---
c:
.- -
CI)
(,)
:=:
-
CI)
(,)
0
4
�
- •
CI)
II) --
--
cv
c:
-
�
- 2
cv
CI)
.s:::.
cv
�
CI)
>
0 0
40 50 60 70 80
E vaporatmg
· temperature ( · c )
Figure 6.1 9 E f f e c t o f the e vaporating t empera ture on the o v e rall
h e a t transfer coefficient for wa ter in a cone e v a porator.
The s olid line is connecting e xperimental values a n d
the d a s h e d line i s connecting theoretical values.
Temperature difference: 1 0K , feed flow: 4 .2x 1 0-s m 3 /s,
rotating speed: 1 57 rad/s.

higher than that in falling film evaporators. It was thought that if the energy of the
vapours could be used to rotate the tubes, the heat transfer coefficient in a falling film
evaporator could be increased to match that of the Centritherm evaporator without

expending extra energy. The vapour velocity down the tubes in a falling film evaporator
is very high, as much as one third the velocity of sound (Jebson and Iyer, 1 99 1). The
kinetic energy of vapours leaving the bottom of the tubes is mostly dissipated, although
some is used to pass the vapours through the vapour-liquid separator. The tubes could
be rotated by mounting a simple bearing at their tops and bottoms, extending the tubes
sightly at the bottom, and cutting and twisting the extended portion so that a simple
turbine is formed. By this means it would be hoped that high speed rotation could be
achieved simply, effectively and cheaply.
In order to test the effectiveness of rotating the tube on the overall heat transfer
coefficient, a motor driven rotating tube falling film evaporator was constructed and
tested. The effect of feed flow, tube rotating speed, temperature difference between
condensing and evaporating temperatures, evaporating temperature and feed temperature

on the overall heat transfer coefficient was investigated.
6.2.3.1 Reynolds number
Figure 6.20 shows the change in the overall heat transfer coefficient in the rotating tube
falling film evaporator with the average liquid film Reynolds number. The Reynolds
number was varied by changing the feed flow. In the figure two regimes can be
identified and these flow regimes are probably related to changes in the liquid flow
pattern, as the Reynolds number is varied. The initial steep rise in heat transfer
coefficients is probably related to a change from laminar to wavy laminar and starts at
a Reynolds number of 30 (Hallstrom, 1 985). Then the heat transfer coefficients increase
slightly with the increase of the Reynolds number. The second regime (flow rate
beyond 0.03 kg/s, corresponding to Reynolds number of 3500) shows the heat transfer
coefficients increase again with increasing of the Reynolds numbers, which is likely to
correspond to a turbulent regime. These changes of the heat transfer coefficient with
4.0
Wavy T urbulent
laminar
�
N
3.5
E
......
3:
�
-
r:::
Q)
.2
-
- 3 .0
Q)
0
0
....
Q)
-
en
r:::
C'O
....
-
- 2.5
C'O
Q)
.s:::.
to
....
Q)
>
0
2.0
0 2000 4000 6000 8000
R e ynolds number
Figure 6.20 Effect o f the R e y no lds number on the overall heat
transfer coef fici e n t o f a rotating t u b e falling film
evapora tor. E v a p o r a ting temperature: 70·C, temperature
difference: 1 0K , r o t a ting spe ed: 7 2 .3 r a d/s, tube
length: 2 m, e v a p o r a ting liquid: w a ter.

the Reynolds number in rotating tube falling film evaporator are similar to the results
obtained in falling film evaporator (Schwartzberg, 1 988). B ut the Reynolds number

where the changes in flow regime are apparently taking place in the rotating tube
evaporator are higher than those observed in a stationary tube (Chen, 1 992). The
rotation of the tube probably favours the changes towards more turbulent regimes at
higher Reynolds numbers. This has been called the re-Iaminar phenomena (Cannon and
Kays, 1 969).
As the rotating speed and feed flow rate were both likely to interact and influence the
flow regime, these factors were studied together using a factorial design with five levels
for the rotating speed and three levels for the feed flow rate.
Figure 6.21 shows the effect of the Reynolds number on the overall heat transfer
coefficient as a function of the rotating speed, which is varied by changing the feed flow
rate. Experimental data shown in Figure 6.2 1 was statistically analyzed and yielded the
regression equation:
Hexp = 2.62 + 0.925 .0 - 0.836 .02 + 0. 1 77 .03 + O.OOO l 04 Re (6-3)

- 0.0004 1 2 .0 Re + 0.000536 .02 Re - 0.000 1 34.03 Re
(R2 = 87.3%, df= 37)
It can be seen in Figure 6.21 that the heat transfer coefficient increases then declines as
the rotating speed increases, and the peak heat transfer coefficient increases with the
Reynolds number. When the tube starts to rotate, the liquid film on the inner surface
of the tube changes the configuration due to tube vibration, which results in better heat
transfer through the liquid film. At low Reynolds numbers, the rise of heat transfer
coefficient occurs at a rotating speed of 1 0 rad/s, then the heat transfer coefficient
decreases as the rotating speed increases. For high Reynolds numbers, the peak heat
transfer coefficient occurs at 80 rad/s, which may be explained as the film thickness is
thicker at high Reynolds numbers (high feed flow rates), and the change of the
4 • Reynolds number: 5370
JJ.. Reynolds number: 1 700
�
N � R eynolds number: 790
E
.....
3= 3.5 •
..:.:
-
c-
CD
0
-
-
CD
0
0 3
"-
-
CD
en
C
IV
"-
-
-
IV
.r=
CD 2.5
-
IV
"-
CD
>
0
2
o 20 40 60 80 1 00
R otating speed ( rad/s )
Figure 6.2 1 E f f e c t o f the tube r otating speed on the overall heat
transfer coefficient i n a rotating tube falling film
e v a p orator at different Reynolds numbers. E v a p orating
temperature: 70·C, tube length: 2 m, tempera ture
difference: 1 0K. e vaporating liquid: water.

configuration will last longer, so the heat transfer coefficient reaches the peak. at a high
rotating speed. However, when the rotating speed is towards the higher limit, a severe
vibration was observed in the experiments. This may result in the detachment of the
liquid film from the rotating surface, consequently reducing the heat transfer coefficients
at high rotating speed. Even if the vibration problem was overcome, the maximum
increase in heat transfer coefficient is likely to be only of the order of 30%. Under
these conditions the use of a rotating tube falling film evaporator is unlikely to be
economic.
Figure 6.22 shows the effect of increasing evaporation temperature on the heat transfer
coefficient. As comparison the results of Chen and Jebson ( 1 992) for a fixed tube
evaporator are indicated in the same figure. The increase is probably due to the changes
in the liquid properties with the temperature. Viscosity will decrease, and thermal
conductivity, and thermal capacity increase, the later only very slightly. The figure also
illustrates clearly that higher heat transfer coefficients are obtained with a rotating tube.
The increase of both heat transfer coefficients for rotating and fixed tube with
evaporating temperature appears to follow a linear relationship being the slope for the
rotating tube higher than that for the fixed tube. The rotating tube at higher evaporation
temperature seems to encourage the film being thinner than the film on stationary tube
at same evaporation temperature.
6.2.3.4 Feed temperature
The effect of the feed temperature for both fixed and rotating tube evaporators is
illustrated in Figure 6.23. This figure shows that the heat transfer coefficient increase
slightly with increase in feed temperature into the evaporator. A similar effect was also
observed in a falling film evaporator (Chen, 1 992) and the results are also plotted in
Figure 6.23.
3.75
• Rota ting s p e e d:
3.5 74.35 r a d / s
•
.A Rotating s p e e d:
-
3.25 o rad/s
c::
CI)
.�
-
-
CI)
o
o
3. 0
-
co
CI) 2.75
.c
co
....
CI)
>
o
2.5
40 50 60 70 80 90 1 00
E vaporatmg
· temperature ( · c )
Figure 6.22 Effect of t h e evapora ting temperature on the overall heat
transfer coe fficient in a rotating tube falling film e v apora tor.
Temperature d i f f e rence: 1 0K , feed flow: 5x 1 0- 3 m 3 1 s,
tube length: 2 m . evaporating liquid: wa ter.

3.5 • R o ta ting speed: 72.3 rad/s
f e ed flow: 3.3x 1 0- s m 3 / s
A R o t a ting speed: 0 rad/s

�
N
E
...... f e e d flow: 2.3x 1 0- s m 3/ s
3=
�
3
-
c
...
•
Q)
•
..
0
00-
00-
CD
0
0
....
Q)
00-
c
en
I'CI
2 .5
....
-
-
I'CI
CD
.r=
....
cu
Q)
>
0
2
SO 70 80 90
F e e d temperature ( · C )
Figure S.23 Effect of the feed temperature on t h e overall h e a t
transfer coefficient i n a r otating t u b e falling film
evaporator. Temp erature difference: 1 0K, evaporating
temperature: 7 0·C, tube length: 2 m. e vaporating
liquid: water.
With the short tube length used in these experiments, the entrance effect will have a
significant effect on the heat transfer coefficient for the tube as a whole. As the feed
temperature increases, while the evaporating temperature keeps constant, there is more
flashing and higher vapour velocities at the top of the tube. The normal flow regime
is then achieved more quickly. This consequently will cause an increase in the overall
heat transfer coefficient. However, when the tube is rotating the centrifugal forces may
also encourage the flow to reach the normal regime fast, and the effect of the vapour
velocity at the top of the tube will become relatively smaller. Hence the effect of
increasing the feed temperature on a rotating tube evaporator will be smaller.
An experimental design with five levels of rotating speeds and three levels of
temperature differences was used to determine the effect of rotating speed on the overall
heat transfer coefficient as a function of the temperature difference.
Figure 6.2 shows the results of the experimental design. The data shown in Figure 6.24
was statistically analyzed giving the following regression equation:
Hexp = 4. 1 9 - 0. 1 34 �T + 0.499 Q - 0.0493 Q3 (6-4)
(R2 = 94.4%, df = 4 1 )
It is noted that in this case there is no interaction between the temperature difference
and the rotating speed, and hence the maximum occurs at a constant speed of 75 rad/s.
Chen and Jebson ( 1 992) showed that for a falling film evaporator there is a rapid linear
fall in the heat transfer coefficient with increase in temperature differences up to 8K,
after which the heat transfer coefficient decreases more slowly. The change in the slope
of the line is thought to be associated with the onset of nucleate boiling. Figure 6.24
shows that for a small temperature difference there is a rapid increase in the heat
5
• 6. T: 4 K
• 6. T: 1 0 K
4 .5 T 6. T: 1 5 K
�
N
E
.......
� 4
.¥
-
c:
Q)
0
:;:: 3.5
-
Q)
0
0
....
-
Q)
en 3 •
c:
....
«I •
-
-
«I
s::.
Q)
2.5
-
....
«I
Q)
>
0
2
0 20 40 60 80 1 00
Rotating speed ( rad/s )
Figure 6.24 E f f e ct o f the tube r o t a ting speed on the o v er all heat
transfer coefficient in a rotating tube f alling film
evaporator a t different temperature differences. Feed
f l o w rate: 5x 1 0-s m 3/s, tube length: 2 m, evapor ating
temperature: 60 · C . evaporating liquid: water.

transfer coefficient with the commencement of rotation, but after that the increase is
slower. In this case it may be attributed that even a low rotating speed there is a rapid
decrease in film thickness and hence a higher heat transfer coefficient, but higher speeds
are not correspondingly effective in decreasing film thickness. It is noted that at the
higher temperature differences there is no rapid increase in heat transfer coefficient at
low speeds. If the main mechanism at these speeds is ebullition, then the effect of
increasing speed may be to remove bubbles more quickly, and this would explain the
steady increase with increase of rotating speed. The decline of the heat transfer
coefficient at high speeds is again thought to be the result of vibration disturbing the
film.
6.3. Conclusions
A theoretical model that describe the flow of a film on a rotating surface has been
developed. Although the model was simple, it revealed basic relationships among
variables for the evaporating heat transfer of the liquid film on the surface of a rotating
cone.
The theoretical equations obtained from the proposed model show that the liquid film
thickness and the heat transfer coefficients are affected by the feed flow rate, rotating
speed of the cone, length of the cone, angle of the cone, and liquid viscosity.
The local film thickness decreases with decrease in feed flow rate, increase of cone
length and rotating speed. Increase of cone angle results in thinner liquid films and
higher heat transfer coefficients.
The theoretical calculated overall heat transfer coefficients show similar trend to
experimental results obtained using the Centritherm and a constructed cone evaporators.
The discrepancies between experimental and theoretical results could be attributed to the
following reasons:
1 ) It was assumed in the theoretical model that the liquid film moves across the conical
surface as a smooth film without waves. However, it was observed by Shinn ( 1 97 1 ) that
waves exist in the liquid film when it moves across the rotating conical surface. Waves
are likely to increase turbulence and hence heat transfer.
2) It was also assumed in the theoretical model that evaporation takes place only at the
surface of liquid film, which is true only at lower temperature differences. At high
temperature differences, bubbles are likely to be formed on the heating surface, which
will affect the heat transfer.
3) Centrifugal forces acting on the liquid film increase the pressure in the film as the
rotating speed rises. This causes that the saturation temperature of the liquid on the
conical surface increases slightly and therefore reduces the temperature difference
between the conical surface and the liquid film.
4) Differences between calculated and experimental heat transfer coefficients could also
be c aused by the presence of non-condensable gases in the steam jacket, which depend
on the efficiency of non-condensable gases removal by the steam-trap system. The

concentrations of non- condensable gases were not measured but it has been reported that
it could seriously affect the steam side heat transfer coefficients (Mincowycz and
Sparrow, 1966; Machereth, 1 993).
The results obtained in falling film evaporator with rotating tube showed that rotating
the tubes of a falling film evaporator will increase the overall heat transfer coefficient
but the increase obtained is very dependent on feed flow rate, and is not sufficient to
j ustify the use of this evaporator in the industry. Rotation also appears to delay the
onset of bubble formation in evaporation, and reduce the effect of vapour velocity at the
top of the tubes.
PART I I I
FOULING I N EVAPORATORS

Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces 1 23
CHAPTER 7
LITERATURE REVIEWS OF FOULING
IN MILK PROCESSING
In virtually all types of heat transfer plants, fouling of heat transfer surfaces is the main
cause of progressive decline in efficiency and perfonnance of the equipment with time.
To overcome fouling, frequent cleaning of equipment is required. Fouling is of
considerable importance in the dairy industry (Tissier and Lalade, 1 986). The fouling
of a heating surface during the processing of milk, especially at high temperatures, is
an important practical problem, with implications for operating costs and product
qUality. The understanding of the fouling process will assist in the determination of
methods to prevent or delay the undesirable deposit formation on the heat transfer
surfaces, and will be more practically important in the design of an efficient cleaning
system.
After looking at the general aspects of the fouling problem, this review mainly focuses
on fouling in milk processing.
7.1 The costs of fouling
Surveys of fouling in the British industry (Pritchard, 1979) estimated that fouling costs
were about £300-500 million per annum, which was between 0.2 to 0.33% of the Gross
National Product (GNP) of the United Kingdom in 1978. The fouling costs include
more than £50 million per annum for the food industry (Fryer and Gotham, 1 988). A
survey in the United States showed that the annual cost of fouling of heat exchangers
run between US $8 to 1 0 billion per annum (Garrett-Price, 1 985). This represents 0.228
to 0.35% of the American GNP for the year 1 984. A survey of fouling in New Zealand
industry estimated that fouling costs were between 0. 1 % to 0. 1 5 % of GNP in 1988
(Steinhagen et ai., 1 990).
Chapter 7 Fouling in Milk Processing - Literature Review 1 24
These figures show that the costs of fouling in the industry are quite significant.
Therefore, the study of fouling is an important aspect in heat transfer equipment and
process design. However, this is still the major unresolved problem (Whalley, 1 992).
Although the precise costs of fouling are difficult to quantify, it is important to examine
the areas in which fouling affects production (Fryer et ai. , 1 995). The total costs
attributable to fouling in heat transfer equipment are made up of the following
components (Pritchard, 1988; Bott, 1 989, 1 995 and Steinhagen et aI. , 1 990):
( 1 ) Capital costs
Extra heat transfer area is included in the design of plant to take account of the
reduction in heat transfer efficiency due to fouling. In practice, the heat exchangers are
between 10%-50% overdesigned, with the average around 30% (Garrett-Price, 1 985).
In some instances of severe or perhaps unknown fouling potential, the allowance of heat
transfer area for fouling might be as high as 1 00% (Bott, 1 988), or even 500% (Sandu
and Lund, 1982). In addition, extra pump capacity is also installed.
(2) Energy costs

Fouling results in increased energy costs via the extra heat and pumping duties required.
(3) Maintenance costs
This includes the costs for installing complex cleaning processes systems and costs for
chemicals.
(4) Loss of production

The necessity of frequent cleaning of the plant, and operation during production at a
reduced efficiency due to fouling, resu�ts in a loss of production.
(5) Impairing product' s quality

Fouling in the food processing industry can give rise to problems of texture, taste and
sometimes aroma of food products (Bott, 1 988) and has to be regarded as a potential
Chapter 7 Fouling in M i lk Processing - Literatu re Review 125
source of biological contamination (Pritchard, 1 988). The quality of the processed milk
may be impaired by the presence of detached pieces of deposit, which break away from
the vessel walls under the action of high shear forces in the final products (Swartzel,
1 983, Marrs and Lewis, 1 986, Grandison, 1 988a).
7.2 The types of fouling
Fouling is normally classified into following six major categories according to the
underlying process (Hallstrom, 1 988a, Bott, 1 995):
( 1 ) Precipitation fouling
This involves precipitation of dissolved substances and/or formation of crystals on the

surface. If the dissolved substances have inverse solubility the precipitation takes place
on heating surfaces at a higher temperature than the liquid. This type of precipitation
is also called scaling.
(2) Reaction fouling

This type of fouling is the result of one or more chemical reactions between reactants
contained in the flowing fluid at, or near, the surface due to the change in temperature.
(3) Particulate fouling

This is the name given to the accumulation of suspended solid particles from the fluid
stream onto the surface. In a few cases the deposition occurs as a result of gravity, in
which the process is referred as sedimentation fouling.
(4) Biological fouling

This type of fouling is due to the attachment and growth of micro-organisms originating
from the process stream on the surface.
(5) Corrosion fouling
Chemical reactions in the material of the heating surface itself.

Chapter 7 Fouling in Milk Processing - Literature Review 126
(6) Freezing fouling

Solidification of the fluid or a fluid component due to a temperature lower than the
melting point at the surface.
In most cases, more than one type of fouling occurs in the system. However, it is useful
to study the mechanisms separately in order to have some appreciation of the factors
that are likely to influence the fouling problem, although in combined fouling systems
there may be modification of the effects of certain variables (Bott, 1989).
Milk fouling on the heating surface may fall into the first, second and fourth categories.
7.3 The research history of fouling in dairy industry
Fouling problems associated with food processing have been studied in processing
sugar-beet liquor and sugars, fruit and vegetable juices, brewing fluids, soya and egg
proteins (Marrs and Lewis, 1 986; Fryer and Gotham, 1 988). In the dairy industry, its
importance has also long been recognised. The research work started in the early
thirties (Parker and Johnson, 1930). As the processing plant required cleaning before
the deposit layer became severe due to hygienic reasons, the initial research of milk
fouling concentrated on plant cleaning as a practical solution (Johnson and Roland,
1 940a,b; Bell and Sanders, 1 944). Only in the last two decades, the fouling of process
plant had been accepted as a part of a larger problem and studied more widely and in
greater detail (Burton, 1 98 8). This is because that the volume of milk processed in a
dairy plant has steadily increased, as well as the reductions in energy consumption,
amounts of cleaning material used and time spent in cleaning have become considerably
more important (Hege and Kessler, 1986a). The research work carried out at the
National Institute for Research in Dairying (NIRD), Shinfield (UK), started in the mid
sixties, provides a starting point for many other fouling studies in the milk industry
(Fryer et ai., 1 995). Now the literature on milk fouling is extensive and significant
progress has been made in understanding the composition and structure of fouling
deposits as well as the fouling mechanisms. The results of these works can be found
in several conference proceedings (Melo et aI., 1988; Kessler and Lund, 1 98 9 ; Belmar
Beiny and Fryer, 1994), reviews (Burton, 1968, 1 988; Bott, 1 989; Fryer et al. , 1 995,
Jeurnink et ai., 1996) and a book (Bott, 1995).
The composition of the deposits is now known, and the chemical changes that occur
when milk is heated are fairly well understood. However, the interaction between the
chemistry and the engineering factors in processing plant are not, and the progress made
towards the reduction of fouling in commercial processing plant has been largely
empirical (Fryer et aI., 1 995).
The difficulty of interpreting the interrelationships of the chemical reactions, which give
rise to fouling, with the fluid mechanics of the heat transfer equipment is due to a
number of reasons (Belmar-Beiny et al., 1993):
(i) The milk fouling deposit is complex. Deposit is mainly formed by a mixture of
proteins and inorganic salts.
(ii) Fouling from milk shows strong seasonal variations, as a result of differences in its
composition. This makes it difficult to interpret results in terms of basic kinetics unless
a well-defined fluid, which does not show such a variation, is used.
(iii) Fouling is generally considered to be the result of several temperature-dependent

effects; chemical reactions which result in the denaturation of whey proteins and the
formation of insoluble protein aggregates, and solubility limits which result in the
precipitation of salts.
(iv) The milk fouling is also a function of many other factors, such as the fluid velocity,
surface conditions, air content, etc.. Fluid travelling through the heat transfer equipment
may experience a range of shear rates and temperatures, both of which affect the rate
of fouling.
7.4 Composition and structure of milk fouling deposit
The extent of fouling on the heat transfer surface and the composition of the deposit
formed from the milk are largely dependent on the temperature of milk. Two distinct
types of deposits were described by Burton (1968):
(i) The first, called Type A, is a voluminous material, which starts to form at
temperatures above 70°C, is maximum in the range 95 to 1 10°C, and then decreases at
higher temperatures (Burton, 1968; Lalande et ai., 1984). The deposit has a high protein
content (50 to 70%) and a lower mineral content (30 to 40%) (Lyster, 1 965, Lalande
et al., 1 984; Tissier et aI., 1984). At the lower end of the temperature range at which
this deposit is formed, most of the protein is �-lactoglobulin, but at the higher end of
the range it is predominantly casein (Tissier et al., 1 984; Lalande et aI., 1985). This
deposit is predominantly soft, white and spongy: however, if overcooked it can become
brown and very much harder (Lyster, 1965 ; Burton, 1968).
(ii) The second type of deposit, called Type B, forms at higher temperatures than Type
A, above 1 20°C. It is hard and granular, and has a high mineral content (70 to 80%)
and a small amount of protein ( 1 0 to 20%), largely casein (Lyster, 1 965; Burton, 1 968;
Lalande et al., 1984).
Table 7 . 1 gave a detailed chemical analysis of fouling deposits .at pasteurization and
UHT temperatures (Tissier et al. , 1 984). The deposit formed in a pasteurizer is of the
Type A while in a UHT sterilizer is close to Type B deposits. The protein composition
of these deposits is shown in the Table 7.2.
As shown in the tables, the composition of milk fouling deposits is highly temperature
dependent. Type A deposit (72°C) contains mainly whey proteins and Immunoglobulins
while the sterilizer Type B deposit contains only casein. There are large differences
between deposit protein compositions and those of raw milk. Therefore fouling results
from preferential deposition of certain milk components.
--- - --
Table 7.1 The composition of milk deposits in a pasteurizer and a sterilizer (From
Tissier et ai., 1984).
Composition in the deposit Pasteurizer Sterilizer

(%) (72°C) ( 1 3 8°C)
Protein 50 12
Mineral 15 75
Fat 25 3
Minor components 10 10
Total 100 1 00
Table 7.2 The protein composition of milk deposits in a pasteurizer and a sterilizer
(From Tissier et al., 1 984).
Protein composition Pasteurizer Sterilizer
(%) (72°C) ( 1 38°C)
�-lactoglobulin 50 -
a-Casein 18 27
"�-Casein -
50
Immunoglobulins 23 -
Minor components 9 23
Total 100 1 00
Although fat is present in the deposits, it does not play a significant role in the process
of fouling (Maas et al., 1985, Burton, 1 988). It is thought that fat occurs in the deposit
Chapter 7 Fouling in Milk Processing - Literatu re Review 1 30
when it is trapped by other constituents (Burton, 1 988). In directly heated UHT systems
Burton ( 1 968) found that after steam-inj ection heating, a different deposit on the
unheated surface of the holding section of the plant was fonned. This deposit had an
unusually high fat content (30 to 35%), with a mineral composition similar than deposits
produced in indirect systems. Lalande et al. ( 1 984) found a similar type of deposit after
homogenization of milk at 1 38°C in an indirect heat exchanger. Burton ( 1 988) proposed

that this deposit is due to destabilization of the fat globules by mechanical action at high
temperature.
Lactose is hardly found in milk deposits because it is water soluble. So even if it was
incorporated in a deposit, it would dissolve again in the milk stream (Jeurnink et ai.,
1 996).
The fouling material not only varies in composition according to the temperature at the
point where it fonns, but it also varies throughout the thickness of the fouling material
(Burton, 1 988). A number of detailed examinations about the structure of fouling
deposits have been reported (Hege and Kessler 1 986b; Daufin et ai., 1 987; Britten et ai.,
1 988; Foster et ai., 1 989). Tissier and Lalande ( 1 986) reported that, for a process time
greater than about 1 hour, Type A deposits consist of two layers: a protein-rich outer
layer and a sublayer, near to the heat exchange surface, that is rich in calcium and
phosphorus. They ascribed the fonnation of the sublayer to the diffusion and subsequent
crystallization of insoluble calcium phosphate. Due to the distinct two layers structure,
there is an argument about the nature of the initial layer of deposits fonned on the
heating surfaces. More discussions regarding initial deposition are given in the section
7 .9.
Type B deposit is hard, granular and brittle in structure (Lyster, 1965; Lalande et ai.,
1 984). Foster and Green ( 1990) studied the structure using scanning electron
microscopy (SEM), and reported that Type B deposits become rougher as the surface
temperature increases. The depth profile of Type B deposits was also studied; no
distinct structural layers were found, but, like Type A deposits, protein was concentrated
near the outside of the deposit with calcium, phosphorus and magnesium near the heat
exchange surface. In addition, calcium phosphates tended to be located nearer to the
surface than magnesium phosphates.
Lyster ( 1 965) noted that calcium and phosphate typically accounted for 90% of the
mineral content of milk fouling deposit, but only represent 30% of mineral content in
raw milk. The calcium/phosphate ratio for Type A deposit was found to be about 1 .5.
This, which has been also verified by others researchers (Ito and Nakanishi, 1 967;
S kudder et ai., 198 1 ; Lalande et ai., 1984; Tissier and Lalande, 1 986), indicates the
presence of calcium triphosphate.
7.5 Factors affecting milk fouling on the heating surface
Dairy fluid fouling is a function of many variables, including chemical and physical
properties. The seasonal changes in milk composition are also reflected in fouling. The
factors which affect the milk fouling on the heating surface can be classified as: milk
properties; and, plant construction and operation.
7.5.1 Milk properties
7.5.1.1 Compositional variations with season
Milk is a complex biological fluid subject to compositional variations (Walstra and

Jenness, 1 984), which are reflected in fouling. B urton ( 1 968) found that different milk
supplies vary widely in their ability to deposit fouling material during heat treatment.
The amount of deposit formation during milk processing can vary from one milk supply
to another, from day to day with a single milk source, and from season to season for
milk from a single herd supply, although the underlying chemical factors are not very
clear (Burton, 1 966, 1 967, 1 968, 1972, 1 98 8 and Grandison, 1988a,b,c). Grandison
( 1 9 88c), however, was unable to correlate herd diet and fouling deposit.
7.5.1.2 pH values of milk
The effect of milk pH on deposit fonnation has been widely studied (Hege and Kessler,
1 986a; Skudder et al., 1 986; Visser et al., 1986; Chaplin and Lyster, 1 988). Milk pH
typically varies between 6.6 to 6.8. Skudder et al. ( 1 986) showed that the amount of
fouling increased markedly at an acidic pH, but remained unaffected at alkaline values.
These results were consistent with the reports of others (Yoon and Lund, 1 989). Hege
and Kessler ( 1 986a) concluded that the denaturation of protein is accelerated by lower
pH values, because a decrease in pH causes a reduction in the negative charges of the
protein molecules. This in tum reduces the electrostatic (repulsive) force between the
molecules, thus, promoting aggregation.
7.5.1.3 The amino nitrogen level of the milk
A strong positive correlation between depositional rate in a heat exchanger and the
amino nitrogen level of the milk has been shown by Lalande and Corrieu ( 198 1 ). This
may depend on the association of the amino group concentration with urea content
(Burton, 1 988). It is known that added urea increases milk heat stability (Robertson and
Dixon, 1 969; Muir and Sweetsur, 1976) and decreases deposit fonnation.
7.5.1.4 Calcium phosphate
Burton ( 1 968) hypothesized firstly that mineral deposition occurred because of the lower
solubility of calcium phosphate at higher temperatures. This view was supported by
Visser et al. ( 1 986). A precipitation fouling was suggested for calcium phosphate
deposition on the heating surface by Barton et al. ( 1985). The rate of precipitation of
phosphate is a function of the temperature and the resulting degree of supersaturation
(Sandu, 1 989). Jeumink and de Kruif (1995) found that either increasing or decreasing
the calcium concentration in respect to standard calcium concentration in milk led to
more fouling, due to the effect of calcium on the stability of casein micelles. D aufin
et al. ( 1 987) showed that dairy fluids not containing calcium did not produce any
fouling and when no precipitable calcium was presented, deposit growth was in fact
stopped. These results might be an explanation for the effect of the seasonal variation
of milk composition on fouling and the inhibiting precipitation of calcium phosphate
could be a method to prevent or minimize fouling.
7.5.1.5 Air content in milk
The presence of air in milk has the effect of increasing fouling rates (Gynning et ai. ,
!
1 958; Burton, 1 968; Marrs and Lewis, 1�86; Fryer, 1986). It is thought that air bubbles
are formed at heated surfaces, which promotes protein coagulation and denaturation.
Marrs and Lewis ( 1 986) concluded that the dramatic reduction in fouling rates, which
resulted from an increase in the back pressure applied to the product side of the heat
exchanger, was almost certainly due to the suppression of air bubble formation at the
heated surface. The air content of bovine milk may be increased by entrapment during
milking (Butler, 1 988). Deaeration of the milk prior to heating reduces the degree of
fouling by anything up to 50% (Gynning et ai., 1 958).
The enhancement of fouling through air bubble formation on the heated surface was
regarded as a distinctive mechanisms of milk fouling (Jeumink et ai., 1 996). Visual

observations showed that air bubbles were nuclei for the formation of deposit (Jeumink,
1 995a). The fouling process induced by the air bubbles can be described as follows:
when the bubbles are formed, the surface becomes dry (Visser, 1 988) and as a
consequence, there is an increase in the temperature difference between the hot stainless
steel surface and the bulk of the liquid, resulting in evaporation of water at the vapour
liquid interface. Due to this evaporation milk is transported from the bulk to the surface
where the air/vapour bubbles are attached. Here milk protein accumulates and because
of the local increase in concentration and the high temperature the protein may coagulate
and deposit on the surface. Eventually the air/vapour bubble bursts and part of the
membrane is carried away with the liquid. The schematic representation of this process
is shown in the Figure 7 . 1 . The contribution of air/vapour bubbles to the deposit is
determined by the amount of air present in the milk, the temperature difference between
Milk flow --.-
. --+
Bubble membrane
fOImed
o
o
o
Bubble busted
Figure 7.1 The schematic presentation of the milk fouling process induced by an
air bubble at a hot stainless steel surface (From Jeumink, et ai.,
1 996)
the surface and the bulk, the operational pressure in the heat exchanger and the wall
shear stress (Jeumink, 1 995a). It is also thought the similar phenomena may occur in
film evaporators for dairy products as a result of nucleate boiling (Jeumink et ai., 1 996).
7.5.1.6 Concentration of milk
The depositional rate increased with the rising of the milk concentrations. Fryer et ai.
( 1 992) found that the total amount of deposit from whey protein concentrate increased
linearly with protein concentration. It was deduced that the increased concentration
resulted in higher viscosity and low turbulence, an increased probability of non-uniform
heating, low heat transfer rates which caused an increased product depositional rate
(Jelen, 1 98 1 ).
7.5.2 Plant construction and operation
7.5.2.1 Storage time
Storage of milk at room or refrigerated temperature for 12-24 hours prior to processing,
without change in pH, significantly reduced deposit formation (Burton, 1966, 1 968;
Ashton, 1 966), although longer-term aging led to a rise in deposit formation (Burton,
1 968; Jeumink, 1 99 1 ). The reason for the initial drop of deposit may result from the
lipolysis of the milk fat by natural lipases in the milk. One of the fatty acids (capric
acid) liberated during lipolysis was found to have a specific. effect, at very low
concentrations (about 1 mgllitre), of reducing the amount of deposit on the heated
surface (Al-Roubale and Burton, 1 979). The late rise of deposit after longer storage
may be due to the action of proteolytic enzymes, secreted by pschrotrophic bacteria
(species growing at temperatures of 5 to l OOC) in the milk. lC-casein is hydrolysed by
proteolytic enzymes during storage, resulting in a decreased heat stability of casein
micelles, which will coagulate on heating and cause an additional protein deposition
(Jeumink, 1 99 1 ) . If milk cannot be effectively refrigerated, bacteriological spoilage may
occur in milk. Therefore, a compromise will be required between beneficial effects of
aging and the adverse effects of bacterial growth and a drop in pH.
Chapter 7 Fouling in Milk Processing - Literatu re Review 136
7.5.2.2 Preheating
Preheating milk was found to reduce fouling in later processes (Bell and S anders, 1 944).
Foster et al. ( 1 989) found that preheating of whole milk at 70 to 95°C for 5 minutes
caused a 10 to 20 % decrease in the amount of deposit produced at the heating
temperature of 100°C, compared with raw milk. It is now known that the denaturation
of whey proteins, especially B-Iactoglobulin, during preheating and preholding stage,

reduces the formation of proteinaceous deposits in later processes. It was also thought
that the supersaturation and crystallization of calcium and phosphorus from solution at
high temperature was responsible for the effect of preheating (Burton, 1 968). Proper
selection of preheating conditions (temperatures and preholding time) could be used as

a method to reduce fouling in the following heat treatment steps. It was also suggested
that in order to minimize the deposition in the holding section, the holding vessels
should be high-volume one, which give a high volume/surface ratio, i.e. reducing the
surface area (Burton, 1 98 8 ; Jeurnink, 1 995b), or a continuous stirred tank reactor rather
than a plug flow reactor or a holding tube (de long et al., 1992).
7.5.2.3 Absolute liquid velocity
In general, the higher the liquid flow velocity, the less the amount of fouling. This is
probably because at lower velocities the thickness of the laminar sublayer adjacent to
the heat surface is greater, so that the volume of material subject to high temperature,
and the volume which remains for relatively long periods near to the surface, are greater
(Burton, 1988). It was demonstrated by Gordon et al. ( 1 968), Fryer ( 1 986) and Gotham
( 1 990) that both the amount and extent of fouling decreased with increasing flow rate
in a tubular heat exchanger, in which it is clearly possible to relate flow rate to the shear
stress in the equipment. It was found that deposition was prevented if the wall shear
stress exceeded 15 Pa when heating simulated milk ultra filtrate in a rotating disk device
(Jeumink et aI. , 1996). It is much more difficult to relate fouling to flow rate in plate
heat exchangers because of the complex flow geometries involved (Fryer et al., 1995).
It was pointed out by B ott ( 1 989) that the benefits of reducing fouling at high liquid
flow velocity must be balanced against the additional pressure drop and associated
pumping costs.
7.5.2.4 Surface conditions
As the milk fouling occurs on the solids and liquid interface, the roughness of the
surface could be important in the fouling. Rough surfaces, in general, appear to favour
deposit growth (Bott, 1 989). It is conceivable that modification of the surface could
prevent or limit fouling. Much theoretical attention, e.g. coating the surface, high polish
of the surface, and electrochemical modification, has been given to this possibility
(Baier, 1 98 1 ), but so far with little practical effect (Burton, 1988). It was also found
that after the first deposit layer is established, the surface roughness will no longer affect
the growth of deposition (Dupeyrat et ai., 1987; Britten et ai., 1 988; Yoon and Lund
1 989). This is not surprising, as once a surface has become covered with fouled
material it is the deposit-deposit interaction, rather than the deposit-metal surface

interaction, which is the most significant factor. Britten et ai. ( 1 988) found that
different coatings on the heating surface did not affect the amount of deposit formed,
but did affect the strength of adhesion; the interfacial energy parameters of the surface
appeared to be the main factors affecting the adhesive strength. They concluded that
the polar binding potential of the coated surfaces was the main factor influencing the
strength of adhesion. It thus appears that the first material to adsorb is responsible for
the strength of the deposit-surface bond; this may affect the cleaning rate (Fryer et ai.,
1 995).
The precleaning conditions should be given major attention before operations are started,
because the predeposited dirt on the surface will act as kernels for secondary deposition
(Visser, 1 988).
7.5.2.5 Temperature
Temperature directly detennines the severity and type of fouling on the surface. As the
temperature rises both on the heat transfer surface and in the liquid stream, the whey
proteins start to denature, aggregate and deposit and the calcium phosphate may start
to precipitate. If temperature increases further, the colloidal stability of casein micelles
in milk decreases, then the coagulation of casein micelles may be become a major
controlling factor in the deposition process. Table 7 . 1 demonstrates that at 7 5°C, heat
denatured B -Iactoglobulin accounted for the majority of the protein in the deposit,
whereas at 1 38°C, B- and a- caseins were present in greater quantity. This is

presumably because the B-Iactoglobulin is denatured at temperature of 75°C, while the
casein micelles become destabilised at higher temperatures.
7.5.2.6 Processing time
In practice, the amount of fouling deposit always increases with time. Fouling in the
dairy industry is more severe than that in other industries. For example, in the
petrochemical industry, it is common to clean only once a year or less (Crittenden et ai. ,
1992) . In comparison to the dairy industry, it is necessary to clean the processing plant
daily or even more often.
The fouling process is a transient process. If fouling resistance (Rt) is plotted against
processing time, different fouling curves may be produced depending on the conditions.
Figure 7.2 summaries the ideal curves that are possible in fouling process (Sandu and
Lund, 1 982; B ott, 1 988, 1 995):
Curves A and A' represent linear fouling processes, in which the masses of deposit
gradually increase with time but the rates of deposition are constant. Curves B and B '
denote the asymptotic fouling processes, in which the rates of fouling for w hatever
reasons, gradually fall with time and eventually reach steady state when there is no net
Chapter 7 Fou l ing in Milk Processing - Literature Review 1 39
c·
B'
Possible
induction
Time
period
Figure 7.2 Idealised fouling curves

increase of deposit on the surface. Curves C and C' represent falling rate fouling
processes.
For Curves A ' , B' and C' , the fouling processes are further complicated by the fact that
a period time has to elapse before significant fouling begins. This period is referred to
as the induction period and usually attributed to the time required to "condition" the
surface (Bott, 1988). This phenomenon is often observed in food fouling and has
resulted in two-stage model for the food fouling process being proposed (Fryer, 1989).
In tubular geometries an induction period, which is characterised by unchanged
conditions of heat transfer and pressure drop, is seen (Delsing and Hiddink, 1983),
whereas in plate heat exchangers the induction period is less obvious due to its complex
geometry (Fryer and Gotham, 1988). In a recent experimental work using a plate heat
exchanger and at heating temperature below 95°C, Delplace et al. ( 1 994) did find that
fouling by whey protein solutions is a two-stage phenomenon.
7.6 Possible mechanisms for milk fouling on the heating surface
A large number of studies were focused on the fouling mechanisms using a variety of
different dairy fluids, processing conditions and analytical techniques.
In general, fouling is a process that can be also described in terms of colloid chemistry.
Since in almost any conceivable case, the nature of solid surface will be different from
those of the contaminating materials, initial fouling can be described by a process of

heterocoagulation. Once the first layer has been deposited, a second layer can be
formed, either by homocoagulation when identical particles deposit onto the already
deposited ones, or by a second heterocoagulation process, when "other" particles are
attracted by and bound to the initially deposited material (Visser, 1 988).
General considerations for a possible mechanism of milk fouling process were proposed
as follows (Burton, 1 988):
( 1 ) Converting--one or more of the constituents in the product is converted into a :fr:rm
capable of being deposited on the surface;

(2) Transporting--transfer of the depositable material (foulant or foulant precursor) to
a surface;
(3) Absorption--fouling material attached to the surface to form an initial layer;
(4) Build-up--deposition of further fouling material on the initial layer;
(5) B alance--between deposition and removal through the shear forces caused by t h e
flow of the products across the deposit-liquid interface.
More recently Visser et al. ( 1 995), from their experimental results, proposed a similar
but more specific milk fouling process which was distinguished by four independent
steps.
( 1 ) The formation of particles in the bulk solution generated by heat;
(2) The deposition of these particles onto the heating surface through diffusion;
(3) The adherence of particles;
(4) The conversion of the adhering particles into a two dimension gel when the surfu::e
concentration is sufficiently high.
There is general agreement that at least two separate reactions occurring during heating
of milk play an important part in fouling: denaturation and aggregation of whey protein,
especially B-Iactoglobulin, and the precipitation of calcium phosphate from solution
(Burton, 1 988). A schematic presentation of fouling mechanisms during heating whey
and milk (below 1 00°C) was proposed by Jeumink et al. (1996) and shown in Figure
7.3.
The whole process can be described as follows: At room temperature a monolayer of

protein is immediately absorbed. Further deposition of protein on top of this monolayer
occurs if whey protein undergoes heat denaturation. The active B -Iactoglobulin
molecules and aggregates are formed in the bulk and transported to the surface. Their
deposition is through reactions with already deposited molecules and this process is
enhanced in the presence of calcium. Calcium phosphate may precipitate directly on the
WHEY
active active
o
�- l g hea' . 0 � _-.;a:.:S;.:.SOC=l.::..at:.:l.=;on-=--
2S·C J 1( �2 ') � 6T
I] QJJJJ�t 11ln 1,essrti1

� s tee 1 . 1IIII1I1I1II1n /f/f/ftn
MILK
111111111111111117�7IJI"7""I/ 111111
Figure 7.3 Schematic presentation of the fouling mechanisms during
heating of whey and milk. ( From leurnink et al. ; 1 996)

stainless steel wall, promoted by a large temperature difference between the heating
surface and the bulk, or it may associate with B-Iactoglobulin aggregates. In milk,
where casein micelles are present, calcium phosphate may also associate with the
micellar calcium phosphate and the active B-Iactoglobulin molecules may associate with
the lC-casein at the surface of the casein micelle and may so entrap the micelles in the
deposit.
In milk fouling, other reactions, such as reactions of lactose with the formation of
organic acids and the removal of carbon dioxide which result from changes in pH of the
solution, may also be involved and their effect in causing the fouling appears to be
interrelated with the deposition mechanisms discussed above (Fryer et ai. , 1 995).
7.7 The initial deposited layer of milk fouling
In order to understand the fouling process, many studies were focused on the initial
events occurring in milk fouling. It is thought that if the processes which govern the
initial period of deposition were understood, it might be possible to extend induction
period and thus improve the industrial heat transfer perfomance. Therefore the sequence
and the rate of events which make up the initial stage are very important (Fryer et al.,
1995) .
Although i t is well known that the fully-developed deposit consists o f a thin mineral-rich
layer adjacent to the metal surface, there is still an argument about the nature of the
initial layer adsorbed on the metal surface. The described fouling mechanisms proposed
by leurnink et al. ( 1 996) was based on results that as soon as a stainless surface comes
into contact with a whey protein solution, even at room temperature, immediately a
monolayer of protein adsorbs. However, Burton ( 1988) described that the first species
formed at the heating surface are mineral crystal nuclei. This deposit is formed due to
the supersaturation of the mineral salts produced by the higher temperature in the thin
liquid layer adjacent to the surface. The results of Daufin et al. ( 1 987) and Green et al.
( 1 988) suggest that mineral deposition i s crucial i n the initial stage of fouling and the
minerals are the most likely species to adsorb fIrst.
Many other workers' results support the theory that proteins adsorb first at the surface.
B aier ( 1 98 1 ) used internal reflectance spectroscopy to study a system at 50°C and found
that proteins were the first species to adsorb in milk fouling. Delsing and Hiddink
( 1 983) (bulk temperature of 76°C), followed by Daufin et al. ( 1 987) (bulk temperatures
from 25 to 88 °C), found, using X-ray photoelectron spectroscopy (XPS) and different
test fluids, that proteins were the main, though not the only, species to adsorb fIrst.
Tissier and Lalande ( 1 986) did not find any minerals in a deposit of 0.0 1 69 f.IIIl
thickness formed after 1 min. However, when using X ray microanalysis they found
high concentrations of calcium and phosphorus at the interface.
Belmar-Beiny and Fryer ( 1993) used X-ray photoelectron spectroscopy to study the
deposits formed by whey protein concentrate after 4 and 1 50 seconds and found that the
composition of deposit film after 4 seconds was close to the theoretical composition of
pure B-Iactoglobulin. Only after 1 50 seconds of contact time were sulfur and calcium
detected; no phosphorus was found. These results indicate that proteins are most likely
to be the first species to adhere to the stainless steel surface. Proteins are very surface
active, and a clean metal surface has a large free surface energy gradient so it would be
expected that a layer would form very rapidly. The role of calcium and phosphate ions
in milk and whey fouling is quite complex. In the bulk, calcium ions promote
aggregation of B-Iactoglobulin not only by binding to the B-Iactoglobulin molecule but

also by decreasing the denaturation temperature (Xiong, 1 992); on the surface, calcium
will form bridges between adsorbed protein and the protein aggregates formed in the
bulk and then adhering to the surface (Fryer et al., 1995 ) .
7.8 The rate-controlling processes i n milk fouling
From the proposed milk fouling mechanisms, it can be considered that the fouling rate
is controlled by two processes, i.e. the mass transfer of active molecules and aggregates
Chapter 7 Fouling i n Milk Processing - Literature Review 145
to the surface, and the chemical reactions, which involves a surface reaction and a bulk
reaction. The slowest of these processes will be the rate-controlling process. Based on
these hypothesis, three cases can be considered (Fryer et aI., 1 995):
(i) If fouling is mass transfer controlled, the transfer of reacted protein to the wall is the
slowest step, and the rate of deposit formation will not be a strong function of
temperature.
(ii) If fouling is controlled only by surface processes, deposition will occur wherever the
wall temperature is hot enough for protein denaturation and aggregation to occur,
regardless of the bulk temperature, the fouling should depend only on the wall
temperature.
(iii) If the controlling reaction for fouling takes place in the bulk, then two conditions
can be envisaged:
(a) If the wall temperature is such that protein denaturation and aggregation will occur,
but the bulk temperature is such that native protein is thermally stable, fouling will
result from deposition of protein which has denatured and aggregated in the thermal
boundary layer adjacent to the wall.
(b) If both the boundary layer and the turbulent core are hot enough for protein
denaturation and aggregation, protein denatured and aggregated in both regions will
contribute to deposit formation.
Gotham et aI. ( 1 989) identified two regimes of protein deposition during the fouling
period in a tubular heat exchanger with a 35% whey protein solution; fouling from the
thermal boundary layer (controlled by surface reaction at the wall temperature) and
fouling from the bulk of the fluid (controlled by bulk reaction, mass transfer and
adhesion of active protein aggregates). In the former regime, the rate and extent of
fouling depends on the wall temperature regardless of bulk fluid temperature, while in
the later regime, fouling occurs when the bulk milk becomes hot enough to produce
denatured and aggregated proteins.
Gotham ( 1 990) demonstrated that when whey protein concentrate passes through a
tubular heat exchanger with the same temperature difference, an increase in fouling
occurred when the bulk temperature exceeds 7 5°C, which suggested that the bulk
reaction of whey proteins and mass transfer is the controlling process. B radley and
Fryer ( 1 992) also showed that, when the bulk fluid is hot enough to produce denatured
and aggregated proteins, fouling deposits will increase when mass transfer to the surface
is increased due to a higher denatured protein concentration gradient.
B elmar-Beiny et al. ( 1 993) outlined two simple models, based on wall and bulk
reactions, that demonstrate that wall react�ons alone cannot be responsible for fouling.
It is not proposed that bulk processes are the only ones involved, or that they will be
important in all situations; the final stage in fouling must be the incorporation of deposit
into the surface, which must involve a surface reaction.
The above results suggest that fouling due to milk fluids is not purely controlled by
surface reactions but that bulk reaction of milk fluid and mass transfer may be involved,
depending on the configuration of the system and its operating variables, such as
temperature and flow rate.
7.9 Whey proteins and their denaturation
From a colloid chemistry point of view, the colloidal components in milk, casein
micelles and fat globules, are sufficiently stabilised by double-layer forces and hydration
layers so that fouling problems at temperature below 100°C are mainly related to its
soluble components, i.e. whey proteins, soluble casein and calcium phosphate (Visser,
1 988). Therefore it is necessary to know the details of the chemistry of whey proteins
as well as their denaturation processes.
Whey proteins comprise about 20% of the protein in bovine milk. They are mainly
composed of the 4 compact globular proteins: B-Iactoglobulin (B-Ig), a.-lactalbumin (a.
la), bovine serum albumin (BSA) and several classes of immunoglobulin (Ig).
Comprehensive reviews concerning the whey protein system of bovine milk are
available (Eigel et al., 1 984; Mulvihill and Donovan, 1987; Fox, 1 989).
7.9.1 fi-lactoglobulin (fi-lg)
B-Ig represents about 50% of the total whey proteins in bovine milk or 1 2% of the total
milk proteins. A molecule of B-Ig consists of 1 62 amino acid residues, has a monomeric
molecular weight of about 1 8 ,300 Da (Eigel et ai. , 1 984) and exhibits well developed
secondary, tertiary and quaternary structures. The secondary structure of bovine B-Ig
consists of about 45% B-sheet, 1 0% a-helical and 45% unordered structures (Timasheff
et al., 1 966, 1967, Creamer, et al., 1983). Depending on the pH, B-Iactoglobulin
molecules may exist as a monomer, dimers or octamer. Below pH 3. 5 , the dimers

dissociate to monomers due to strong electrostatic repulsion forces. In the pH range
from 3.5 to 5.2, the dimers combine to form octamers. Between pH 5 . 5 and 7.5, the
molecules exist as stable dimers. Above pH 7.0, the molecule undergoes reversible
conformational changes, and above pH 8.0, the protein is unstable and forms aggregates
of denatured protein (Lyster, 1 970). It was also found that the monomer-octamer
association is not only affected by pH, but also by factors such as protein concentration,
temperature and ionic strength (Langton and Hermansson, 1992). The quaternary
structure is maintained largely by electrostatic forces (Jelen and Rattray, 1 995).
7.9.2 a-lactalbumin (a-Ia)
a-Ia is a small protein molecule with a molecular weight of 14,200 Da. It comprises
about 20% of the whey proteins, and is generally reported as being the second most
important protein. The molecule consists of 1 23 known amino acid residues (Fox, 1 989;
Walstra and Jenness, 1 984). It is a very compact molecule and nearly spherical in
shape. Its secondary structure consists of 26% a-helix, 14% B-sheet and 60% unordered
structure at physiological pH (Robbins and Holmes, 1 970). a-Ia normally exists as a
monomer.
7.9.3 Bovine serum albumin (BSA)
BSA is a major component of blood serum, but is also found in whey. It is synthesized
in the liver, and goes into the milk via the secretary cells. It represents about 10% of
the whey proteins, contains 1 7 disulphide bonds and one free thiol group per molecule
(Reed et ai., 1 980). This holds the molecule in a multiloop structure. B S A consists
of 582 amino acid residues and has a molecular weight of about 66,000 Da (Peters,
1985). The secondary structure of BSA consists of about 10% B-sheet, 5 5 % a-helical
and 25% unordered structures (Reed et aI., 1 975).
7.9.4 Immunoglobulins (lgs)
Igs represent 1 0% of the whey proteins. Igs are polymers of two identical light chains
and two identical heavy chains joint together by disulphide bridges. They are far the
heaviest proteins in whey with the molecular weight of 1 .5 x 105 - 9 X 1 05 Da (Jelen
and Rattray, 1 995). The Igs are among the most heat-sensitive of the whey proteins
(Fox, 1 989).
7.9.5 Whey protein structure
Whey proteins in their native state are globular proteins having secondary and tertiary
structures. The three dimensional globular structure of the individual whey proteins are
stabilised by intramolecular disulphide bonds, hydrogen bonds, hydrophobic interactions
and ionic bonds.
The maj ority of the non-polar hydrophobic amino acid side chain residues are on the
interior, and the polar hydrophillic residues are on the exterior part of the molecule.
The intramolecular bindings and a loosely held water layer around the" molecules make
interactions with other molecules difficult (Clark et aI., 1 98 1 ).
7.9.6 Whey protein denaturation and aggregation
Protein denaturation is a process in which the native protein structure changes without
alterations in the original amino acid sequence. Thennal denaturation is due to a sudden
entropically induced disruption of the native state (Clark et al., 1 98 1 ). When heated
above about 70°C, the structure of B-Iactoglobulin is irreversibly altered by two separate
processes of denaturation and aggregation. These processes are consecutive; the native
confonnation of the protein first is partially unfolds into disordered and non unifonn
molecules. This is an intramolecular process which exposes the hydrophobic molecular
core, together with reactive disulfide and sulfuydryl (- SH) bonds, to an aqueous
environment. This unstable configuration is then stabilized by polymerization with other

denatured molecules in intennolecular aggregation that may lead to bind the protein
molecules to dimers, quatromers, and oligomers and further on to gels (Clark et al.,
1 98 1 , Hege and Kessler, 1986a). Although denaturation is reversible at low

temperatures, as molecules can return to their folded state on cooling, aggregation of
whey proteins is totally irreversible; the resulting aggregates are insoluble in water
(Fryer et ai. , 1 995). The rate of reaction increases significantly between 7 1 and 75°C
(Donovan and Mulvihill, 1 987). One difficulty in interpreting the literature in this area
is that the two processes are very difficult to separate, and the whole process is often
described as denaturation (Fryer et ai., 1 995).
7.10 The relationship of whey protein denaturation and milk. fouling
Although the exact mechanisms of milk fouling are still not very clear, the correlation
between protein denaturation and fouling of heat exchangers has been confirmed by
many investigators (de Jong et ai., 1 992, 1 993 ; Belmar-Beiny et ai. , 1 993). Results
obtained by Tissier et al. ( 1 984) showed that B-Ig has an important role in the fouling
process of heat exchangers. It appears that denaturation of B-Ig and the fonnation of
deposits evolve in parallel along the heat exchanger. Even for whipping cream the
correlation between the denaturation of whey proteins and the fouling rate has been
demonstrated (Hiddink et ai. , 1 986). Hegg et al. ( 1 985), working with solutions of
whey proteins, showed that deposit formation and denaturation was greatest at the
isoelectric point of the proteins. They therefore suggested that there was a close
connection between protein denaturation and deposit formation.
The measurements of Hegg and Larsson ( 1 98 1 ) have shown that in milk, B-Ig is most
prone to deposit on metal surface, whereas a-Ia has a lesser tendency to attach to
surfaces.
Hege and Kessler ( 1 986a) pointed out that both the effect of temperature and the
activation energy for the process of deposit formation are similar to that on whey protein
denaturation. Kessler et al. (1986) found that the activation energy for demineralized
whey fouling was 3 1 5 kllmol, whereas the activation energy for protein denaturation
was 300 kJ/mol. However, Fryer ( 1 986) and Lalande and Corrieu ( 1 98 1 ) found lower
values for milk fouling: 87 kllmol and 90 kJ/mol respectively. It is likely that different
controlling reactions have different rate-limiting steps and therefore different activation
energies.
Gotham et al. ( 1 992) suggested that aggregation rather than denaturation of whey
proteins is the controlling step in deposit formation. Jeurnink ( 1995b) concluded from
his experimental results that the denaturation of the whey proteins, specially the
formation of the active B-Ig molecules and aggregates, in the vicinity of the surface is
a prerequisite to obtain deposition.
7.11 Conclusions
This review shows that the fouling in milk processing has been studied considerably, but
most of the research work was concentrated on tubular or plate heat exchangers and
UHT plants. There is lack of knowledge of fouling in other processing units. For
example, fouling in evaporators has received little attention. It was known that fouling
in multistage evaporators were severe at both first effect and last effect, which were due
to the high temperature in the first effect and the high concentration in the last effect.
Little infonnation regarding fouling in the evaporators is available in the published

literature (Kessler, 1 989; Jeurnink and Brinkman, 1 994). The objective of this research
work is to study the fouling in an evaporator with a rotating surface.

CHAPTER 8
FOULING IN EVAPORATORS - MATERIALS AND METHODS
8.1 Introduction
The study of fouling in the evaporation process is limited. No publication of fouling
in the Centrithenn evaporator is available in the literature at present. The objectives of

this part of the work were to investigate the fouling phenomena occurring in a
Centrithenn evaporator and to evaluate the effect of operation variables on the onset and
rate of fouling.
8.2 Materials
8.2.1 Equipment
( 1 ) A Centrithenn Evaporator (CT 1B-2, ALFA-LAVAL) and a rotating falling film

evaporator were used;
(2) A temperature recorder (Monitech) ;
(4) A refractometer (ATAGO, No.25870, Japan) ;
(5) A PAGE system (Mini protein II equipment, Bio-Rad Laboratories, Richmond,
USA) ;
(6) A pH meter (PHM 6 1 Laboratory pH meter, Copenhagen, Denmark)
(7) A drying oven (Clayson, Catalogue O I l 24, Laboratory apparatus Ltd, Upper Hutt,
New Zealand);
(8) Other instruments used include an analytical balance, a stop watch and a measuring
cylinder.
8.2.2 Model solutions
To avoid compositional and seasonal variations in feed material, two model feeds were
first used in the preliminary trials.
Chapter 8 Fouling in Evaporators - Materials and Methods 153
( 1 ) Low heat skim milk powder from Tui Milk Products Company was used to make
reconstituted skim milk solutions at 10, 20 and 25% TS.
(2) Powdered whey protein concentrate (ALACEN 1 34) from the New Zealand Dairy
Board was used to make up a reconstituted whey solution at 26% TS.
The solutions were made by sprinkling weighed powder onto a weighed quantity of
water (to avoid the formation of agglomerates) in a locally designed milk powder mixer,
which was available in the pilot plant at the Food Technology Department, Massey
University. Mixing was continued for 5 minutes after all the powder was added, to
make sure that any agglomerates or any powder sticking to the walls of the container
were dissolved. The solutions were then stored in a chill room before use.
As no detectable fouling was found in the preliminary trials, a sweet cheddar cheese
whey solution (total solids = 6.5 - 7.5%, pH = 4.4 - 4.8) from a local cheese factory
(Tui Milk Products Company) was chosen for the remainder of the experimental work.
All the experimental work was carried out within a two months period in the spring.
Thus the compositional, seasonal and daily variations of whey solutions could be
regarded as negligible. Manufacture of the cheddar cheese was normally done in the
early morning (4-5 am), therefore the whey solution was collected in plastic milk cans
and stored in a chill room (4°C). The whey then was transferred to the pilot plant at the
Food Technology Department and stored in a chill room until the experiment was
carried out.
8.2.3 Chemicals used for cleaning
The caustic alkaline solution, the acid solution and the sanitiser solution used in this part
of experimental work were the same as described in section 4.2.4.

Chapter 8 Fouling in Evaporators - Materials and Methods 1 54
8.3 Methods
8.3.1 The Centritherm evaporator and the falling film evaporator
The experimental work was mainly carried out on the Centritherm evaporator system.
which was described in section 4.3. 1 . Some experimental work was also conducted on
a falling film evaporator, which was described in section 4.3.2. A comparison of the
results obtained with these two evaporators were made. The falling film evaporator was
operated without rotation.
8.3.2 The recycle system
One problem with fouling experiments is that good quality milk is relatively stable in
the evaporator and the decline in heat transfer coefficient will be detected only after a
quite long time (several hours), which means that experimental work on milk fouling
is time-consuming and expensive. Alternative approaches to understanding milk fouling

have focused on breaking milk down and examining whey protein concentration, or
permeates. These solutions foul at a much faster rate than milk, and provide an insight
into the fouling process although they may have different fouling characteristics to milk
(Kastanas et ai., 1 995).
To obtain a proper picture of the fouling process in an evaporator, the processing time
has to be 4 to 6 hours or even longer. This necessitates the use of large amounts of
feed material, which incurs in high costs and practical difficulties to carry out the
experiment in a pilot plant. Therefore, to minimize raw material utilisation, a recycled

whey solution was employed in this experiment. About 40 litres of whey were recycled
In each trial.
To recycle the whey solution in the system and maintain a constant concentration of the
feed, the concentrated solution and the vapour condensate were measured and fed back
to the feed container.
When using whey with this recycling system, it was found that the fouling resistance
would reach a constant value after running for 4 hours. Therefore, the rest of runs were
stopped after this time.
8.3.3 The examination of denatured whey proteins
Polyacrylamide Gel Electrophoresis (PAGE) was used to determine the extent of whey
protein denaturation during processing in the Centritherm. The gel preparation and the
procedures of using PAGE are given in the Appendix VI.
8.4 Instrumentation
This part was the same as that described in section 4.5.
8.5 Experimental
8.5.1 Experimental variables
The levels of experimental variables used are shown in Table 8 . 1 .
Table 8.1 Ranges of experimental variables
Evaporating Temperature (ET) 60, 70°C
Temperature difference (dT) 1 0, 20K
Rotating speed (RS) 1 05, 1 86 radls
Feed flow (FF) 1 .67x lO-5, 4.2x l O-5 m3/s

Chapter 8 Fouling i n Evaporators - Materials and Methods 1 56
8.5.2 Experimental design
A 42 factorial experimental plan was employed using the Plackett-B urman experimental
design program (supplied by the Food Technology Research Centre, Massey University)
to find out interactions between variables.
Table 8.2 The arranged trials with selected variables.
Trial ET �T RS FF
1 -
- -
-
2 -
- + +
3 - + - -
4 - + + +
5 + - - +
6 + - + -
7 + + -
-
8 + + + +
Note: '+' represents a high value, '-' represent a low value
Two trials of the middle point and duplicate of each trial were made.
To assess the effect of individual variables, other trials were also carried out.
8.5.3 Experimental Procedure
The use of the Centritherm evaporator and experimental procedures were the same as
those described in sections 4.7 and 4.8. One modification was made to the experimental
procedure for the measurement of vapour condensate flow. As the fouling process is
a transient process, the measurement of vapour condensate flow rate was done as quick
as possible. Therefore flow was measured for a two minute period (triplicate) at every
hour during each test run.
8.5.4 Evaporator Cleaning
A special attention was given to the cleaning procedure described in section 4.7.2.
An extended long time cleaning procedures was employed in order to achieve maximum
removal of organic or inorganic deposits so that no deposit nucleation sites remain,
which might accelerate subsequent deposition.
After the completion of the trial, the flow was switched to water and the flow rate was
increased to the maximum possible, until the water coming out from the evaporator
becomes clean, which took about two minutes. The flow were then switched from water
to detergent. Two kinds of detergents were used: an alkaline-based solution (AC-1 80)
and an acid-based solution (AC-300). The action of alkaline detergent lasted for 30
minutes, at a concentration of 1 % (w/w), to remove the deposited proteins, then acid
detergent, at concentration of 1 % (w/w), was circulated for 30 minutes to remove the
deposited minerals. Between the detergent stages, and at the end of cleaning, rinsing
with water stages, each lasting 10 minutes, were used.
The cleanliness of the surface was checked once by disassembling the apparatus. It was
observed that the deposits had been removed completely after the cleaning using the
above procedures. B efore each test, the heat transfer coefficient with water was
measured to check that the heat transfer coefficient had not changed, to ensure that the
surface was clean.
8.6 Theory and data processing
The equation for the overall heat transfer coefficient in an evaporator with a: clean
surface can be written as:
1 (8- 1)
Ho
If the effect of fouling is taken into account, a fouling resistance (�) can be included
in the equation as:
(8-2)
or
1 (8 -3)
H(t)
where
(8-4)
0it) is the thickness of the fouling layer at time t.

kd is the thennal conductivity of the fouling layer.
The fouling resistance can be calculated from the overall heat transfer coefficients using
the following equation:
1
R/t) = (8-5)
H (t)
The overall heat transfer coefficients (Ho and H(t» are computed by using the measured
flow rate of vapour condensate (Wexp) , the temperature difference between the steam
condensing and the liquid evaporating temperatures (LlT), and the latent heat (hrg), which
is a function of the evaporating temperature. The calculation formula is the same as the
Equation 4-1 given in the Chapter 4. Thus, the fouling behaviour in the evaporator can
then be estimated by measuring the change in the overall heat transfer coefficients with
time.
Changes in the flow rate of vapour condensate which result from fouling were
monitored in order to determine the changes of overall heat transfer coefficient and the
fouling resistance changes with the time.
All collected data were loaded into the Quattro program. The calculated results are
given in the Appendix VII.
CHAPTER 9
FOULING IN EVAPORATORS - RESULTS AND DISCUSSION
The experimental work presented in this chapter deals with the fouling in a Centrithenn
evaporator under different operating conditions using whey solutions as the fouling
material. The effects of variables, such as evaporating temperature, difference between
the liquid evaporating and the steam condensing temperatures, and rotating speed of the
heating surface on fouling are discussed. The interaction of the experimental variables
on the evaporator surface fouling are also discussed. Possible mechanisms of whey
fouling on the rotating surface are then proposed.
9.1 Preliminary experimental results
Initially, several exploratory experiments were carried out, using reconstituted skim milk
solutions with concentrations of 10%, 20% and 25% total solids, and a reconstituted
whey powder solution with a concentration of 25% total solids, at evaporating

temperatures of 50, 60 and 70°C, a temperature difference of 20K, a rotating speed of
186 rad/s, and a feed flow rate of 4.2 x 1 0-5 m3/s.
A summary of the experimental results are presented in Table 9. 1 . As can be seen in

this table, there was no fouling detectable for these solutions after running for more than
6 hours. These results were not expected. One possible reason for this results is that
materials which are responsible for forming the deposits on the surface have been
removed (inactivated) or reduced during the manufacture of powders. Analysis of
fouled deposits in evaporators has shown the deposits were mainly whey proteins and
minerals (Jeumink and Brinkman, 19 94). The depositable materials, therefore, could
include some minor constituents in their native form (Burton, 1 988), minerals and
denatured whey protein. It has been found that aggregated whey proteins are less active
in the fouling process (Jeumink, 1 995a). Jeumink ( 1 995b), similarly, found that
reconstituted milk gave much less fouling than fresh milk in tubular heat exchangers.
Chapter 9 Fouling in Evaporators - Results and Discussion 161
It was suspected that the process of reconstitution presumably caused other changes
affecting the fouling behaviour.
Table 9.1 Summary of preliminary experimental results after 6 hours evaporation
running
Trials Solutions ET ST RS FF Time Final �

(OC) (OC) (radfs) (m3/s) (hrs) (m2K/kW)
x 10-5
1 1 0% TS Skim
milk 50 70 1 86 4.2 6 0
2 20% TS Skim
milk 60 80 1 86 4.2 8 0
3 25% TS Skim
milk 70 90 1 86 4.2 7 0
4 25% TS WPC 60 80 1 86 4.2 7 0
Another possible reason is that the liquid evaporating temperature and the temperature
of the heating surface in evaporators is much lower, when compared with that in
sterilizers and UHT plants. Thus, there is no fouling to be detected on the surface of
the Centrithenn evaporator by using these solutions.
Hence, reconstituted skim milk and whey solutions, which are frequently reported in the
literature for fouling studies in sterilizers and UHT plants, were not used, instead, fresh
sweet cheese whey was used for the remainder of this study.
C hapter 9 Fouling in Evaporators - Results and Discussion 1 62
9.2 Effect of evaporation temperature and temperature difference on fouling
Figures 9. 1 and 9.2 show the fouling resistance as a function of time at different
temperature differences, feed flow rate and evaporating temperature. The figures clearly
illustrate that fouling resistance increases with time for cheese whey at both evaporation
temperatures 60°C and 70°C (Figure 9. 1 and 9.2 respectively).
All the fouling curves depicted in Figure 9. 1 (evaporating temperature 60°C) show an
induction period less than one hour. However there is no induction period for an
evaporating temperature of 70°C and a temperature difference of 20K. An induction
period of about half an hour was observed for an evaporating temperature of 70°C and
a temperature difference of 10K. It appears that this period was needed to achieve
sufficient denaturation of whey protein to induce fouling.
It was demonstrated by other investigators (Jeurnink, 1 995a, b; de Jong et aZ. ; 1 992,

Hege and Kessler, 1 986a) that the denaturation of the whey protein is strongly
associated with the milk fouling and that the rate of whey protein denaturation increased
as the temperature rose.
As illustrated in the figures there is a strong influence of the temperature difference on

the evaporator fouling, whereas the influence of the feed flow is much weaker. If
Figures 9. 1 and 9.2 are combined for a given feed flow and temperature difference, it
can be clearly seen that evaporation temperature increases fouling rate (Figure 9.3).
In general, the higher the evaporation temperature and temperature difference, the
greater the fouling on the surface. This can be attributed to the fact that depositable
materials adsorb on the surface more quickly at a high temperature than at a lower
temperature (Hegg et al., 1 985; Roscoe and Fuller, 1994).
Chapter 9 Fouling in Evaporators - Results and Discussion 1 63
0.06
-
3=
..:s:.
"'"
�
N
-
E 0.04
CD
()
c:
co
-
CI)
CI)
CD
a:
0.02
C)
c:
::I
0
LL
0.0
o 0.5 1 1 .5 2 2.5 3 3 .5 4 4 .5 5
Time ( h o u r s )
Figure 9 . 1 F o u ling r e s i s t a n c e a s a f u n c t i o n o f time for r e c y c l e d
s w e e t c h e e s e w h e y i n t h e C en t r i t h e r m e v a p o r a t o r
at an e v a p o r a t i n g t e m p e r a t u r e o f 6 0 'C and a
r otating s p e e d o f 1 0 5 r a d / s .
A Tem p erature difference: 2 0 · C. f e e d flow: 1 .67x 1 0-5 m3/s
• Tem p era ture difference: 2 0 · C . feed flow: 4.2x 1 0-5 m3/s
* Tem p erature di fference: 1 0 ·C. feed f l ow: 1 .67x 1 0- 5 m3/ s

Chapter 9 Fouling in Evaporators · Results and Discussion 1 64
0.0 8
3=
.::tt:.
......
� 0.06
N
-
E
CD
0
c:
C'tJ 0.04
-
CI)
CI)
CD
a:
C')
c:
0.02
:::l
0
L1..
0.0
o 0.5 1 1 .5 2 2.5 3 3 .5 4 4.5 5
Time (hours)
Figure 9 .2 Fouling r e s i s t a n c e a s a function o f time for r e c y c l e d
s w e e t c h e e s e w h e y i n the C e n t r i t h e r m e v a p o r a t o r
at an e v a p o r a t i n g t e m p er a t u r e o f 7 0 · C and a
rota ting s p e e d o f 1 0 5 r a d / s .
... T e m p e r a t u r e d i f f e r e n c e : 2 0 · C , f e e d f l o w : 1 .67x 1 0- 5 m 3 / s
* Tem p erature d i f f e r e n c e : 20·C, f e e d f l o w : 4 .2x 1 0- 5 m 3 / s
5 m3 / s
C Tem p erature dif f e r e n c e : 1 0 · C , f e e d f l o w : 1 .67x 1 0-
· 5
o Tem p erature dif f e r e n c e : 1 0 C, f e e d f l o w : 4.2x 1 0- m 3 / s
0.08
3=
..:.=
......
� 0.06
N
-
E
Q)
0
c:
co 0.04
-
en
en
Q)
a:
en
c:
0.02
:J
0
U.
0 .0
o 0 .5 1 1 .5 2 2.5 3 3.5 4 4.5 5
Time (hours)
Figure 9.3 F o u l i n g resis tan c e a s a f unction o f time at
differ en t e v a p o r a t i n g t e m p e r atur e s f o r r e c y c l e d
s w e e t c h e e s e w h e y i n t h e Centri t h e r m e v a p o ra tor.
R o t a t i n g s p e e d : 1 0 5 rad/ S , feed f l o w : 1 .6 7 x 1 0- 5
m3/ S , t e m p e r a t u r e d i f f e r e n c e : 2 0 · C .
• E v a p orating T e m p e r a tu r e : 70·C
• E v a p o r ating T e m p e r a tu r e : 6 0 ·C
The change in the state of protein in cheese whey solutions after running (recycled) in
the evaporator at 60°C and 70°C for 6 hours was detennined using PAGE. It was found
that there were no detectable denatured proteins at 60°C, and only loss of BSA was
found at 70°C.
Table 9.2 Percentage loss of native whey protein during evaporation at a
temperature of 70°C and a temperature difference of 20K
Time (hour) 1 2 3 4 5 6
a-la 0 0 0 0 0 0
f3-lg 0 0 0 0 0 0
BS A 44% 50% 61% 72% 72% 72%
Results for the more severe conditions are presented in Table 9.2. The gel photograph
of PAGE is given in the Appendix VIII .
By using equations derived in Chapter 5, the inner surface temperature of the rotating
cone was calculated to be about 77°C, when the evaporation temperature was 70°C, the
temperature difference 20K, the rotating speed of cone 1 86 rad/s, and the feed flow 4.6
x 1 0-5 kg/so This reveals that although the steam temperature was 90°C, the temperature
at which the whey solutions were subjected was not very high. At this temperature only
BSA would be expected to denature. This explains why only the BSA was 72%
denatured but other whey proteins were undenatured after running the evaporator at 70°C
for 6 hours.
Although the content of BSA in the whey is small, a thin layer of denatured BSA could
be easily attached on the surface and be associated with other depositable materials
existing in the whey. Most of the research work regarding fouling in milk processing
was found that other proteins mainly contribute to deposits. One example is given in section
7.4. When a pasteurizer was operating at noc, only B-Ig, Igs, and a-Casein were included
in milk deposits. Given that the temperature difference used in pasteurization is usually
higher than that for evaporation, the temperature of the steel surface is appreciably higher in
pasteurization. Hence in this process denaturation of j3-lg, Igs, and a-Casein, may occur as
is shown in Table 7 1 (P 129), but is much less likely in evaporation. BSA has not been
.
literature as a precursor of fouling. The results obtained in this work show that BSA
plays an important role in the initial stage of fouling, especially for fouling in the
evaporation process.
9.3 Visual observation of the whey deposits on the surface.
After a run was completed, the Centritherm evaporator was disassembled and the fouled
cone was carefully taken of the machine and visually observed. A photograph of the
evaporator cone is presented in Figure 9.4. Visual inspection showed that the deposited
film was thin and soft, which appears to be of the type A milk deposit described by
other investigators (Burton, 1968, Lalande et aI. , 1 984, Tissier et ai., 1 984). It was
planned to weigh the mass of deposited materials and make a chemical analysis of the
deposit, but it was very difficult to remove a deposit sample from the cone surface. It
was also not practical to insert a test section, which could remove a sample after the
experiment in order to examine the fouled surface. Therefore only a visual observation
was done.
If the kd is taken as 0.3 W/m.K (Wood, 1 982), from a theoretical calculation, the fouled
deposit thickness on the cone surface, assuming a fouling resistance of 0.04 (m2K1kW),
could be determined as 12 1JlIl. Obviously, the low conductivity of the fouling layer
makes such a thin deposit layer to have a significant reduction in the overall heat
transfer coefficient.
9.4 Effect of the rotating speed on fouling
The effect of rotating speed on the fouling resistance at two different feed flows and
temperature differences can be observed in Figure 9.5. It appears that the increase of
rotating speed reduces fouling at low temperature differences. This effect, however, is
not significant at higher temperature differences. This result suggests that temperature
is more important than the liquid flow velocity. It may be explained that at higher
temperature difference, the wall temperature (temperature differences were altered
F ig u r e 9.4 The fou l ed rot a t i n g su rface of the C e n t r i t herm evaporator

0.08
3:
.::.:.
......
� 0.06
N
Q)
u
c:
10 0.04
-
U)
U)
Q)
a::
C)
c:
0.02
::J
0
u..
o 0.5 1 1 .5 2 2.5 3 3.5 4 4.5 5 5 .5 6
T i me ( h o u r s )
Figure 9.5 F o u l i n g r e s i s t a n c e a s a f u n c t i o n of time f o r
recyc l e d s w e e t ch e e s e w h e y i n t h e C e n t r i t h e r m
e v a p o r a t o r a t an e v a p o r a t i n g t e mperature o f
7 0 D C a n d d i f f e r e n t r o t a ti n g s p e e d.
* � T: 2 0 · C , FF: 4 .2x 1 0- s m 3 / s , R S : 1 05 r a d / s
... � T : 2 0 · C , FF: 4 .2x 1 0- s m3 / s , R S: 1 8 6 r a d / s
o � T: 1 0 · C , FF: 1 .67x 1 0-s m 3/ s , R S : 1 0 5 r a d / s
¢ � T: 1 0 · C , FF: 1 .6 7 x 1 0-s m 3 / s , R S : 1 8 6 rad/s

by changing the steam temperature) is higher and activated molecules are formed at the
liquid film close to the wall and attached to the deposited layer, whereas, at lower
temperature difference, the activated molecules are fonned in the bulk solution and then
transferred by mass transfer to the surface, which is affected by the liquid film velocity.
It can also be seen that the final measured fouling resistances after 4 hour operation
depended of the temperature differences.
Figure 9.6 shows a comparison between fouling in the falling film evaporator and the
Centrithenn evaporator. It can be seen that the fouling rate is higher in the falling film
evaporator than that in the Centrithenn evaporator. This shows that the high velocity
of liquid film delays the fonnation of the deposit on the heat transfer surface. In theory,
the higher the rotating speed, the higher the centrifugal force applied on the liquid film.
Under the action of a higher centrifugal field, the velocity gradient in the liquid film
becomes large and the thickness of the laminar sublayer is reduced. Therefore, the
deposition rate is delayed and the increasing rate of thennal resistance with time will
be lower. Taborek et al. ( 1 972) also found that increased flow velocity results not only
in higher heat transfer coefficients, but in a decreased tendency towards fouling.
One point which should be taken into account is that the same amount of whey solutions
were used in each trial for the falling film evaporator and the Centritherm evaporator,
but the heat transfer area in the falling film evaporator was more than the double that
in the Centrithenn (0.2 m2 in the falling film evaporator and 0.09 m2 in the
Centrithenn). Therefore, the final fouling resistance corrected by area in the falling film
evaporator . (about 0. 1 6 m2K/kW) will be much higher than that in the Centritherm
evaporator (about 0.047 m2K/kW).
9.5 The interaction of variables on the fouling resistance
The fouling process is a transient process. To find out the interaction between variables,
the values of fouling resistance after 4 hours operation, obtained in the 4 2 factorial
experiment, were loaded into a regression analysis program (Quattro).

Chapter 9 Fouling in Evaporators ' Results and Discussion 171
0.1
-
0.08
3:
�
......
�
N
E
0.06
CI.>
u
c:
IV
-
en 0.04
en
CI.>
a::
tn
c:
0.02
::1
0
Ll..
0.0
0 0 .5 1 1 .5 2 2.5 3 3 .5 4 4.5 5 5 .5 6
Time (hours)
F i g u r e 9 .6 C o m p a r i s o n of f o u l i n g r es i s t a n c e s in t h e f a l l i n g
fi lm e v a p o r a t o r a n d t h e Cen t r i t h e r m e v a p o r a t o r
f o r s w e e t c h e e s e w h e y a t a temp e r a tu r"e
difference of 20·C.
• E T : 7 0 · C , FF: 4 .2 x 1 0- 5 m 3 ! s (Falling f i l m e v a p o r a t o r )
* E T : 7 0 · C , F F : 4 .2 x 1 0-5 m 3 ! s , R S : 1 8 6 r a d / s ( C e n trith e r m )
4 E T : 6 0 · C , FF: 1 .6 7 x 1 0- 5 m 3 I s ( F a l l i n g f i l m e v a p o r a to r )
• E T : 6 0 · C , F F : 1 .6 7 x 1 0-5 m 3 / s , R S : 1 0 5 r a d / s ( C e n trith e r m )
Chapter 9 Foul ing in Evaporators - Results and Discussion 1 72
The following equation was obtained:
� = 3.4 x 1 0 -3 T - 3 . 1 x 1O-5 ,Q + 5.2 x 1 O 2 Q

evp
-
, (9- 1 )
- 7.5 x l 0-4 T Qf + 1 .4 x l O-6� T,Q
evp
(R2 = 95.0%, df = 1 0)
It was found that the interaction between evaporating temperature and feed flow rate
causes a decrease in the fouling resistance. A possible explanation may be that the
higher evaporating temperature causes a reduction on the viscosity of the evaporating
solution, which in turn encourages turbulence and decreases the fouling rate (Taborek
et ai. , 1972).
It was also found that the interaction between the temperature difference and rotating
speed causes the fouling resistance to increase. This would be showing that the effect
of the temperature difference on the fouling rate is larger than that of the rotating speed.
9.6 The building up of the deposits
During the experimental trials, an interesting phenomena was found. For the same
operation conditions fouling rate increased when a new whey solution was introduced
in the evaporator. It can be seen from Figure 9.7 that the fouling resistances increase
with time after new whey solutions were introduced into the system. The whole process
for the first whey solution was run for the first four hours and then second and third
whey solutions were introduced and run for another three hours respectively. These
results would indicate that depositable materials in the whey would be used up or
inactivated after running for some period and the deposit rate largely depends on the
concentration of the depositable materials (activated molecules). This result agrees with
the fouling model proposed by Jeurnink ( 1 995b). He found that the fouling rate is
strongly related to concentration of activated molecules (e.g. unfolded whey protein) in
the solutions. Therefore, the results of this research seem to indicate that the
denaturation of whey proteins in the vicinity of the surface is a prerequisite to obtain
fouling.
Figure 9.7 clearly illustrates that there is an induction period after a new whey solution
is introduced. However, the induction period decreases, i.e. it is about one hour for the
first solution and about half an hour for the next introduced solutions. This may be due
to the fact that it takes less time for depositable materials to be adsorbed on the fouled
or unclean surface than on the clean surface (Burton, 1 988). So the first induction
period is longer than the following induction periods.
These results suggest that the formation of the activated molecules is a critical factor to
influence the fouling process. At the higher temperature, the activated molecules formed
fast and the rate of deposition was therefore quicker, while at lower temperatures, the
activated molecules fonned at relatively slow rate.
9.7 The possible fouling mechanism of whey solutions on the rotating surface
The above results suggest that the fouling mechanism is due to the fonnation of
activated molecules and it is a critical factor to influence the fouling process. The
formation of activated molecules, largely due to the denatured whey proteins (in this
experimental work, B S A was found as a major activated molecule), is a combination of
time and temperature. At high temperatures, the activated molecules form fast and the
rate of deposition is, therefore, quicker, while at low temperatures, the activated
molecules form relatively slowly and the rate of deposition is also low. The increase
of liquid velocity over the surface delays the formation of the deposits at low
temperatures, which results in a low concentration of activated molecules.

0.1
3:
� 0 .0 8
.......
�
N
E
0.06
Cl)
o
c
�
....
en
en 0.0 4
Cl)
a:
0)
c:
:::l 0.02
o F r e s h whey s o l ut i o n
LL.
intr o d u c e d
0.0
o 1 2 3 4 5 6 7 8 9 10
Time ( h o u r s )
F i g u r e 9.7 F o uling r e s i s t a n c e a s a f u n c t i o n o f time for r e cy cl e d
s w e e t c h e e e s e w h e y , when a n e w w h e y s o l u ti o n i s
i n t r o d u c e d . R o t a ting s p e e d : 1 0 5 r a d / s , e v a p o r a t i n g
temper a t u r e : 5 0 · C , temp e r a t u r e d i fferen c e: 2 0 · C ,
f e e d f l o w : 4.2x 1 0- 5 m 3 / s .
9.8 The evaporation mechanism on a fouled heat transfer surface
In order to study the effect of deposited layer on the overall heat transfer coefficient, a
fouled heat transfer surface (whey solution was run more than 5 hours until a constant
fouling resistance was achieved) was then used to find out how the overall heat transfer
coefficient changed with the temperature difference. The results are depicted in Figure
9.8.
Figure 9.8 shows that the temperature difference affects the overall heat transfer
coefficient after a deposit layer was established on the surface. The same trend was
found for both the fouled and the clear surfaces. It can be then assumed that a similar
mechanism of evaporation exists on fouled and clean surfaces.
9.9 Conclusions
The results obtained can be summarised as follows:

( 1 ) Fouling is strongly linked with evaporation temperature and temperature difference
because of the denaturation of whey proteins, which are temperature dependent.
(2) BSA was found to be an important depositable material involved in the fouling
process with whey solution.
(3) The presence of denatured whey proteins In the vicinity of the surface is a
prerequisite to obtain deposition. Aggregated whey proteins have less ability of
depositing onto the surface.
(4) A small amount of denatured whey protein could reduce the heat transfer coefficients
significantly.
(4) Fouling is also a function of liquid velocity. The effect is more significant at lower
evaporation temperature. Increasing the rotating velocity will delay the fonnation of an
initial layer and reduce the rate of fouling.
(5) Fouled and clean surfaces have the similar evaporation mechanism.
-
�
N
10
E
.... ...
3:
.::t:.
-
c 9
Cl)
0
-
-
Cl)
0
0 8
...
Cl)
-en
c
10
...
- 7
-
10
Cl)
.c
10
... 6
Cl)
>
0
0 10 20 30 40
T e m p er a t u r e d i f f e r e n c e ( 'c )
Fi gure 9.8 O v e r a l l h e a t .. t r a n s f e r c o e fficient a s a f u n c t i o n
o f t e m p erature d i f f e r e nc e for d i f f e re n t l i q u i d s
and clean and fouled rotating surfaces in the
C e n t r i t h e rm e v a p o r a t o r . R o ta t i n g s p e e d : 1 8 6 r a d / s
• W a t e r on cl ea n s u r f a c e , ET: 6 0 ' C , FF: 5 x 1 0- 5 m 3 / s
A S u g a r s o l ution ( 2 0 %) on the c l e a n s u r f a c e ,
� W h e y so lution on f o u l e d surf ac e, E T : 6 5 'C,
FF: 4.2x 1 0- 5 m 3 / s
CHAPTER 10
OVERALL DISCUSSION AND RECOMMENDATIONS
10.1 Overall discussion
10.1.1 Background of the Ph.D. programme
The starting point of this research work was to evaluate the feasibility of a new on-farm
evaporation system, in which a small scale evaporator could be used to pre-concentrate
milk on farms.
The proposed on-farm evaporator consisted of a rotating heating surface to improve heat
transfer and at the same time compresses the vapour. Thus, an understanding of the
factors controlling heat transfer coefficients on the rotating surface was an important
step for the design of the on-farm evaporator and constitute one major part of this Ph.D.
thesis.
The initial idea was to develop a highly efficient falling film evaporator with rotating
surfaces. This was based on the fact that the Centritherm evaporator (rotating surface
evaporator) could provide very high heat transfer coefficients (of the order 10 kW/m2K) .
If this could be achieved in the falling film evaporator with rotating tubes, the efficiency
of the evaporator could be improved significantly. After a series of trials on a falling
film evaporator with a rotating surface, it was realized that the cone angle was a crucial
factor in the Centrithenn evaporator design to produce the high heat transfer coefficient.
Then a cone evaporator (with a 1 0° of half top angle) was built to study the effect of
the cone angle on the heat transfer coefficient in a rotating surface evaporator. At same
time, another idea was generated to develop an evaporator in which the evaporator body
is connected with a vapour compressor. This evaporator was expected to be operated
Chapter 1 0 Overall Discussion and Recommendations 178
in an efficient way by taking the advantages of the high heat transfer coefficient of a
rotating surface evaporator and the nearly isothennal compression of the vapours. A lot
of efforts were made to develop the prototype for this evaporator and a lot of time was
spent to test it. Unfortunately, this idea was not successful as no compressed vapour
was able to be generated.
Then the main attention was focused on the factors affecting heat transfer coefficient in
the rotating surface evaporator and a theoretical model was developed. After that, the
aspect of fouling in this type of evaporator was also investigated.
From this work, a possible design of a suitable evaporator to be used in on-farm
evaporation system has been developed.
10.1.2 Liquid mm flow and heat transfer on the rotating surface
The developed analytical model gave a clear picture of what occurs in a thin film
evaporator with rotating surface in terms of film thickness and heat transfer coefficient.
It is shown that the liquid film thickness and the heat transfer coefficients are affected
by the feed flow rate, rotating speed of the cone, length of the cone, angle of the cone,
and liquid viscosity. The local film thickness decreases with decrease in feed flow rate
and liquid viscosity, increase of cone length and rotating speed. Increase of cone angle
results in thinner liquid films and higher heat transfer coefficients.
10.1.3 Heat transfer in the Centritherm and the cone evaporators
The theoretical calculated overall heat transfer coefficients show a similar trend than the
experimental results obtained in the Centrithenn and the cone evaporators.

In general, the overall heat transfer coefficient increase with the increase of the rotating
speed. Good agreement between theoretical and experimental values of overall heat
transfer coefficients as a function of rotating speed was obtained with the Centritherm
evaporator, when evaporating temperature was 70°C and difference between the steam
condensing and the liquid evaporating temperatures was 1 0K.
The measured overall heat transfer coefficient in the Centritherm evaporator increases
with the increase of temperature difference until 30K and then decreases. This result
can be explained by the fact that the liquid-vapour interface evaporation may occur only
at low temperature differences. Bubble fonnation on the heating surface may occur
when high temperature differences are used. The bubble formation could create
turbulence in the liquid film, which improves heat transfer, but it could also thicken the
film, which reduces heat transfer to the liquid-vapour interface. When the temperature
difference becomes too high, bubbles may start sticking to the surface and reducing the
available surface, which causes the decrease of overall heat transfer coefficient.
The overall heat transfer coefficient increases with increase in evaporating temperature
due to the change in the liquid physical properties, namely the viscosity and thermal
conductivity.
The overall heat transfer coefficient was expected to decrease with increase of the
concentration of sugar solutions, mainly due to the increase of liquid viscosity.
However, it was found that heat transfer coefficients were of the order of 5 kWm2/K for
60% sugar solutions at temperature differences greater than 15K in the Centritherm
evaporator.
The overall heat transfer coefficients experimentally obtained with the cone evaporator
were lower than those measured in the Centritherm evaporator. This shows that the
cone angle plays an important role in the overall heat transfer coefficient of rotating thin
film evaporators. There was a large difference between theoretical and experimental
Chapter 1 0 Overall Discussion and Recommendations 1 80
values in the cone evaporator which may be due to a poor distribution of the liquid mm.
The discrepancies between experimental and theoretical results could be attributed to the
following reasons:
1) It was assumed in the theoretical model that the liquid film moves across the conical
surface as a smooth mm without waves. However, it was observed by Shinn ( 1 97 1)
that waves exist in the liquid fIlm when it moves across the rotating conical surface.
This may be due to the vibration of surface when it rotates. Waves are likely to
increase turbulence and hence the heat transfer.
2) It was also assumed in the theoretical model that the evaporation takes place only
at the surface of liquid mm, which is true only at the lower temperature difference. At
higher temperature differences, bubbles are likely to be formed on the heating surface,
and affect the heat transfer.
3) Centrifugal forces acting on the liquid film increase the pressure in the film as the
rotating speed rises. This causes that the saturation temperature of the liquid on the
conical surface increases slightly and therefore reduces the temperature difference
between the heating surface and the liquid film.
4) Differences between calculated and experimental heat transfer coefficients could also
be caused by the presence of non-condensable gases in the steam jacket, which depend
on the efficiency of non-condensable gases removal by the steam-trap system. The
concentrations of non-condensable gases were not measured but it has been reported that
it could seriously affect the steam side heat transfer coefficients (Mincowycz and
Sparrow, 1 966; Machareth, 1 993).

10.1.4 Heat transfer in falling film evaporator with rotating tube
From the results obtained with the falling film evaporator, it can be concluded that
rotating the tubes of a falling film evaporator will increase the overall heat transfer
coefficient but the increase obtained is very dependent on feed flow rate, and is not
sufficient to justify the use of this evaporator in the industry. Rotation also appears to
delay the onset of bubble fonnation in evaporation, and reduce the effect of vapour
velocity at the top of the tubes.
10.1.5 Fouling with whey solution in the Centritherm evaporator
This work only provides preliminary data on the fouling of the rotating surface during
the evaporation process of fresh whey. It was found that reconstituted skim milk and
whey solutions did not cause any fouling problem after running for more than 6 hours
in the Centrithenn evaporator. This indicates that the aggregated whey proteins, which
were fonned in the manufacture of skim milk powder and whey powder, were less
active to produce fouling.
The results obtained confirmed that temperature strongly influences the fouling
behaviour. In general, the higher the evaporating temperature and the temperature
difference, the fasteris the deposition rate and the greater the fouling on the surface.
Bovine Serum Albumins (BSA) denatured 72% after running the evaporator at a ·
evaporating temperature of 70°C and a temperature difference of 20K for 6 hours.
Though the content of BSA in whey is small, the denatured BSA could be easily
attached to the surface. By associated with other depositable materials existing in the
whey solution, the thin layer of deposit could reduce the heat transfer coefficients
significantly. This was attributed to the lower thennal conductivity of deposit layer.
An induction period was observed when the evaporating temperature was 60°C. It was
also found that the induction period decreased when new whey solutions were
Chapter 1 0 Overall Discussion and Recommendations 1 82
introduced into the evaporator. It proved the fact that depositable materials are much
easier to be adsorbed on fouled or unclean surfaces than on clean surfaces. The results
of increasing fouling rate after new whey solutions were introduced suggested that the
concentration of activated molecules in the solutions strongly affected the fouling
process.
Increasing the rotating velocity would delay the formation of an initial layer and reduce
the rate of fouling. The effect of liquid velocity was more significant at high
temperature than at low temperature.
A possible mechanism for whey fouling on a rotating surface was first the formation of
activated molecules, namely denaturation of proteins (mainly BSA), which is the
prerequisite for the deposition. The concentration of the activated molecules, which
were affected by the evaporating temperature and the temperature difference, determined
the rate of deposition on the surface. When activated whey proteins are aggregated,
they have less ability of depositing onto the surface.
10.2 Recommendation for the future work
( 1 ) A more complicated model using numerical methods could be developed to

describe the evaporation process in rotating surface evaporators. Two particular
aspects, which are the formation of the waves in the liquid film and the wall
superheat required for onset of nucleate boiling in liquid film flow, should be given
consideration.
(2) More experimental work could be conducted to gain a further understanding of the
operation variables, e.g. the effect of changing both the temperature difference and
the rotating speed on the heat transfer coefficient.

(3) Concerning the on-farm evaporation system, which requires a compact evaporator with
high efficiency and minimal damage to milk, a logical development from this research is
to use a rotating surface with an angle close to 90°, which means a disk type evaporator.
It should provide a better heat transfer coefficient and is worth exploring
(4) It is suggested that a detailed investigation on w�ey fouling should be carried out with a
close look at different whey solution components. Analysis of the composition and
structure of deposited material could provide the further understanding of the fouling
process. New techniques, such as X-ray photoelectron spectroscopy (XPS), should be
employed in the further study. A thorough understanding of the mechanism of fouling
should lead to methods to minimise fouling problem.

H eat Transfer and Fouling in Film Evaporators with Rotating Surfaces 1 84
References
AI-Roubale, S. M. & Burton, H., 1979, Effect of free fatty acids on the amount of
deposit formed from milk on heated surfaces, Journal of Dairy Research, 46,
463-47 1 .
Al -Zubaidi, A. A. J., 1 987, Sea water desalination in Kuwait - A report on 3 3 years
experience, Desalination, 63, 1-55, Cited by Mackereth ( 1 993).
Andrews, A. T., 1 983, Proteinases in nonnal bovine milk and their action on caseins,
Journal of Dairy Research, 50, 45-55.
Angeletti, S. & Moresi, M., 1983, Modelling of mUltiple-effect falling-film
evaporators, Journal of Food Technology, 18, 539-563.
Anon, 1 992, Centrifugal-Flow Thin-Film Vacuum Evaporator, Evapor, Enprotech
Corp, New York.
Anon, 1990, Centrifugal, thin-film, vacuum evaporator, The Chemical Engineer,
25( 1 0), 2 1 .
Armerding, G. D., 1 966, Evaporation methods as applied to the food industry,
Advances in Food Research, (Edited by Chichester, O. C.), 15, pp. 303-358,
Academic Press, New York and London,
Ashton, T. R., 1966, Dairy Industries, 31, 480, Cited by Burton ( 1 988).
B aier, R. B., 198 1 , Modification of surfaces to reduce fouling and/or improve
cleaning, Fundamentals and Applications of Surface Phenomena Associated

References 1 85
with Fouling and Cleaning in Food Processing, (Edited by Hallstrom, B . ,
Tragardh, c . , & Lund, D. B.), pp. 1 68- 1 89, University of Lund, Sweden,
Cited by Fryer et al. ( 1 995).
Barton, K. P., Chapman, T. W. & Lund, D. B., 1 985, Rate of precipitation of
calcium phosphate on heated surfaces, Biotechnology Progress, 1, 39-45,
Cited by Fryer et al. ( 1 995).
Bell, K. J., 1 982, Heat exchangers with phase change, Heat Exchangers Theory and
Practice, (Edited by Taborek, J. et al.), pp. 3 - 1 8 , Hemisphere Publishing Co.
New York.
Bell, K. J. & Sanders, C. F., 1 944, Prevention of milkstone formation in a high
temperature-short-time heater by preheating milk, skim milk and whey,
Journal of Dairy Science, 27, 499-504.
Belmar-Beiny, M. T. & Fryer, P. J., 1 994, Fouling and Cleaning in Food Processing,
University of Cambridge, Jesus College, Cambridge, UK.
Belmar-Beiny, M. t. & Fryer, P. J., 1 993, Preliminary stages of fouling from whey
protein solutions, Journal of Dairy Research, 60, 467-483 .
Belmar-Beiny, M. T., Gotham, S . M., Paterson, W. R., Fryer, P . J. & Pritchard, A .
M., 1993, The effect of Reynolds number and fluid temperature i n whey
protein fouling, Journal of Food Engineering, 19, 1 1 9- 1 39.
Benj amin, T. B., 1 957, Journal of Fluid Mechanics, 2, 254, Cited by Fulford ( 1 965).
Bennet, R., 1 996, Personal Communication, Food Technology Department, Massey

University, New Zealand.
References 1 86
Bergles, A . Z. & Rohsenow, W. M., 1 964, The determination of forced-convection
surface-boiling heat transfer, Journal of Heat Transfer, 86, 805-8 15.
Billet, R, 1 989, Evaporation technology: principles, applications, economics, VCH
Ellis Horwood Ltd, Chichester, England and Germany.
Blass, E., 1 979, Gas/film flow in tubes, International Chemical Engineering, 12(2),
1 83 - 1 95 .
Bott, T . R , 1 995, Fouling of Heat Exchangers, Elsevier Science B .V., Amsterdam,
The Netherlands.
Bott, T. R , 1 989, Fouling of heat exchangers: A review, Process Engineering in the
Food Industry - Developments and Opportunities, (Edited by Field, R W., &
Howell, l. A.), pp. 97- 1 09, Elsevier Applied Science, London and New York.
Bott, T. R , 1 988, General fouling problems, Fouling Science and Technology,
(Edited by Melo, L. F., Bott, T. R , & Bernardo, C. A.), pp. 3-14, Kluwer
Academic Publisher�, Dordrecht, The Netherlands.
Bouman, S. & Waalewijn, R, 1 994a, Concentration of dairy products with rotating
evaporators. 1 . Theory and experimental methods, MilchwissenschaJt, 49(4),
1 90- 1 93.
Bouman, S . & Waalewijn, R , 1 994b, Concentration of dairy products with rotating
evaporators. 2. Results and conclusions, MilchwissenschaJt, 49(5), 253-256.
Bouman, S., Waalewijn, R., de long, P. & Van Der Linden, H. 1. L. 1., 1993, Design
of falling-film evaporators in the dairy industry, Journal of the Society of
Dairy Technology, 46(3), 1 00- 1 06.

References 187
Bradley, S. E. & Fryer, P. J., 1992, A comparison of two fouling-resistant heat
exchangers, Biofouling, 5, 295-3 14, Cited by Fryer et ai. ( 1 995).
Brauer, H., 1 956, Chem. -Ing. -Tech., 32, 7 1 9, Cited by Blass ( 1 979).
Briggs, D. c . , 1 959, Heat transfer in rotating turbulent pipe flow, Technic Report No.
45, Department of Mechanical Engineering, Stanford University, Stanford,
California, USA, Cited by Kreith ( 1 968).
Britten, M., Green, M. L., Boulet, M. & Paquin, P., 1 988, Deposit formation on
heated surfaces: effect of interface energetics, Journal of Dairy Research, 55,
55 1 -562.
Brodov, Y. M., Savel ' yev, R. Z., Permyakov, V. A., Kuptsov, V. K. & Gal'perin, A.
G., 1 977, The effect of vibration on heat transfer and flow of condensing
steam on a single tube, Heat Transfer-Sov. Res., 9 (No . 1 ), 1 5- 1 56, Cited by
Carey ( 1 992).
Burton, H., 1 988, Chapt�r 10 Fouling of heat exchangers, Ultra-High-Temperature
Processing of Milk and Milk Products, pp. 292-309, Elsevier Applied Science,
London and New York.
Burton, H., 1 972, Changes throughout lactation in the amount of deposit formation
from milk of individual cows, Journal of Dairy Research, 39, 1 83- 1 87 .
Burton, H . , 1 968, Reviews of the progress o f dairy science: Section G . Deposits from
whole milk heat treatment plant--a review and discussion, Journal of Dairy
Research, 35, 3 1 7-330.
Burton, H., 1 967, Seasonal variation in deposit formation from whole milk on a
heated surface, Journal of Dairy Research, 34, 1 37 - 1 43.

References 1 88
B urton, H., 1 966, A comparison between a hot-wire laboratory apparatus and plate
heat exchanger for determining the sensitivity of milk to deposit formation,
Journal of Dairy Research, 33, 3 17-324.
Campanella, O. H., 1 99 1 , Empirical equation for the viscosity of sucrose solutions,
An unpublished paper, Personal Communication.
Cannon, J. N. & Kays, W. M., 1969, Heat transfer to a fluid flowing inside a pipe
rotating about its longitudinal axis, Journal of Heat Transfer, Transactions of
the ASME, (2), 1 35- 1 39.
Carey, V. P., 1 992, Liq uid-Vapour Phase-Change Phenomena. An Introduction to the
Thermophysics of Vaporization and Condensation Processes in Heat Transer
Equipment, Hemisphere Publishing Corporation, Washington, Philodelphia,
London.
Caric, M. 1994, Concentrated and Dried Dairy Products, VCH Publishers, New
York.
Chavarria, V. M., 1 983, Experimental Analysis of a Single Multi-Tube Absorption
Driven Evaporator, Ph.D. Dissertation, University of Massachusetts, Amherst,
US, Cited by Schwartzberg ( 1 988).
Chen, H., 1992, Factors Affecting Heat Transfer in the Falling Film Evaporator,
Masterate Thesis, Massey University, Palmerston North, New Zealand.
Chen, H. & Jebson, R. S., 1 992, Factors affecting heat transfer in a laboratory falling
film evaporator, Proceeding of IPENZ 1 992 Annual Conference, 1, pp. 44 1 -
446, Christchurch, New Zealand.

References 1 89
Chen, H., Jebson, R. S . & Campanella, O. H., 1 993, Factors affecting heat transfer in
the Centrithenn evaporator, Proceeding of APCChEICHEMECA 1 993
Conference, pp. 455-462, Melbourne, Australia.
Chen, S. L., Huang, K. M. and Hong, 1. T. 1 990, General solution of film
condensation on a rotating cone, Journal of the Chinese Institute of Chemical
Engineering, 2 1 ( 1 ), 37-44.
Chun, K. R. & Seban, R. A., 1 97 1 , Heat transfer to evaporating liquid films, Journal
of Heat Transfer, Transactions of the ASME, 93(4), 391-396.
Chun, M. H. & Kim, K.T., 1 990, Assessment of the new and existing correlations for
laminar and turbulent flow film condensation on a vertical surface,
International Communications in Heat and Mass Transfer, 17, 43 1 -44 1 , Cited
by Machereth ( 1 993).
Clark, A. H., Judge, F. J., Richards, J. B., Stubbs, J. M. & Suggett, A., 1 98 1 ,
Electron microscopy of network structure in thermally-induced globular
protein gels, Intern,ational Journal of Peptide and Protein Research, 17, 380-
392.
Collier, J. G., 198 1 a, Heat transfer in condensation, Two-Phase Flow and Heat
Transfer in the Power and Processing Industries, (Edited by B ergles, A. E. et
al.), pp. 330-365, Hemisphere Publishing Company, New York and London.
Collier, 1. G., 198 1 b, Convective Boiling and Condensation, 2nd edition, McGraw
Hill, New York.

References 1 90
Creamer, L. K., Parry, D. A. D. & Malcolm, G. N., 1 983, Secondary structure of
bovine B-Iactoglobulin B , Archives of Biochemistry and Biophysics, 227, 98-
1 05 , Cited by Jelen & Rattray (1995).
Crittenden, B. D., Kolaczkowski, S. T. & Downey, I. L., 1 992, Fouling of crude oil
preheat exchangers, Trans IChemE, Part A, 70, 547-557, Cited by Fryer et al.
( 1 995).
Daufin, G . , Labbe ' , J. P., Uuemerals, A., Brule' , G., Michel, F., Roignant, M. &
Priol, M., 1 987, Fouling of a heat exchange surface by whey, milk and model
fluids. An analytical study, Lait, 67(3), 339-364.
Davis, E. J. & Anderson, G. H., 1 966, The incipience of nucleate boiling in forced
convection flow, AIChEJ (Am. Inst. Chern. Eng. J.), 12, 774-80.
de Jong, P., Waalewijn, R. & Van Der Linden, H. J. L. J., 1 993, Validity of a kinetic
fouling model for heat-treatment of whole milk, Lait, 73, 293-302.
de Jong, P., Bouman, S. & Van Der Linden, H. J. L. J., 1 992, Fouling of heat
treatment equipment in relation to the denaturation of B-Iactoglobulin, Journal
of the Society of Dairy Technology, 45( 1 ) , 3-8.
Delsing, B. M. A. & Hiddink, J., 1 983, Fouling of heat transfer surfaces by dairy
fluids, Netherland Milk and Dairy Journal, 37, 1 39- 1 48.
Dengler, C. E. & Addoms, J. N., 1 956, Heat transfer mechanism for vaporization of
water in a vertical tube, Chemical Engineering Progress Symposium Series,
52( 1 8), 95- 1 03.
Deplace, F., Leuliet, J. C. & Tissier, J. P., 1 994, Fouling experiments of a plate heat
exchanger by whey proteins solutions, Fouling and Cleaning in Food

---- - ------
References 191
Processing, (Edited by Belmer-Beiny, M. T. & Fryer, P. 1.), pp. 1-8,
University of Cambridge.
Dickey, L. C . & Craig, 1. c . , 1 993, Freeze concentration o f liquid foods, Physical

Chemistry of Food Process: Volume 2, Advanced Techniques, Structures and
Applications, (Edited by Baianu, 1. c., Pessen, H. & Kumosinski, T. F.), pp.
542-553, Van Nostrand Reihold, New York.
Donovan, M. & Mulvihill, D. M., 1 987, Thermal denaturation and aggregation of

whey proteins, Irish Journal Food Science Technology, 11, 87- 1 00.
Dukler, A. E., 1960, Fluid mechanics and heat transfer in vertical falling film system,
Chemical Engineering Progress Symposium Series, 56(30), 1 - 10, Cited by
Fulford ( 1 964).
Dukler, A. E. & Bergelin, O. P., 1 952, Characteristics of flow in falling liquid films,
Chemical Engineering Progress, 48( 1 1 ), 557-563.
Dupeyrat, M., Labbe' , 1. P., 'Michel, F., Billoudet, F. & Daufin, G., 1 987, Wettability
and solid-liquid interaction during fouling of several materials by whey and
milk, La it, 67, 465-486.
Eigel, W. N., Butler, J. E., Emstrom, C. A., Farrell, H. M. Jr., Harwalkar, V. R.,
Jenness, R. & Whitney, R. M., 1 984, Nomenclature of proteins of cow' s milk:
fifth revision, Journal of Dairy Science, 67, 1 599- 1 63 1 .
Elsaadi, M. S., 1992, Heat Transfer to Thin Film Liquids on Closed Rotating Discs
Systems, Masterate Thesis, University of Newcastle, Upon Tyne, U.K, Cited
by Jachuck and Ramshaw ( 1 994).
Eucken, A., 1 937, Naturwissenschaten, 25, 209, Cited by ·Carey ( 1 992).

References 1 92
Ferell, H., Fitch, E. C. & Boggs, J. H., 1 957, An investigation of pressure drop
through a rotating pipe, Am. Inst. Chem. Engrs. Paper 57-PET-9, Cited by
Kreith ( 1968).
Fergusson, P. H., 1 989, Developments in the evaporation and drying of dairy

products, Journal of the Society of Dairy Technology, 42(4), 94- 1 0 1 .
Fluck, A. A . 1 . , 1 988, Priciples o f unit operations: Evaporation, Journal of the

Society of Dairy Technology, 41(4), 94.
Foster, C. L., Britten, M. & Green, M. L., 1 989, A model heat-exchange apparatus
for the investigation of fouling of stainless steel surfaces by milk. I. Deposit
formation at 1 00°C, Journal of Dairy Research, 56, 20 1 -209.
Foster, C. L. & Green, M. L., 1990, A model heat exchange apparatus for the
investigation of fouling of stainless steel surfaces by milk. ll. Deposition of
fouling material at 140°C, its adhesion and depth profiling, Journal of Dairy
Research, 57, 339-348.
Fox, P. F., 1 989, The milk protein system, Developments in Dairy Chemistry. 4.
Functional Milk Proteins, (Edited by Fox, P. F.), pp. 1 -53, Elsevier Applied
Science Publishers, London.
Frost, W. & Dzakowic, A. S., 1967, An extension of the methods of predicting

incipient boiling on commercially finished surfaces, ASME paper 67-HT-61 ,
presented at the 1 967 National Heat Trasfer Conference, Seattle, Washington.
References 1 93
Fryer, P. J., 1 995, Chapter 9 Clean technology in the food industry, Clean
Technology and the Environment, (Edited by Kirkwood, R c., & Longley,
A. J.), Blackie Academic & Professional, London.
Fryer, P. 1., 1 989, The uses of fouling models in the design of food process plant,
Journal of the Society of Dairy Technology, 42( 1), 23-29.
Fryer, P. J., 1 986, Modelling Heat Exchanger Fouling, Ph.D. Thesis, University of
Cambridge, UK, Cited by Fryer et al. ( 1995).
Fryer, P. 1., Belmar-Beiny, M. T. & SchreIer, P. J. R, 1 995, Fouling and cleaning in

milk processing, Heat-Induced Changes in Milk, (Edited by Fox, P.F.), 2nd
Edition, International Dairy Federation, Belgium.
Fryer, P. 1. & Gotham, S . M., 1988, Fouling and cleaning in food processing,
Fouling and Cleaning in Process Plant, pp. 167- 1 8 1 , St. Catherine's College,
Oxford.
Fryer, P. J., Gotham, S. M. & Paterson, W. R, 1 992, The concentration dependence

of fouling from whey protein concentrates, Proc. 20th Austr. Chemical
Engineering Conference (CHEMCA 92), 1, pp. 368-375, Canberra, Australia.
Fulford, G. D., 1 964, The flow of liquids in thin films, Advances in Chemical
Engineering, (Edited by Drew, T. B., Hoopes, J. W., Vermenlen, T. &
Cokelet, G. R), 5, pp. 1 5 1 -236, Academic Press, New York and London.
Fulford, G. D., 1962, Gas-Liquid Flow in an Inclined Channel, Ph.D. Thesis,

University of Birmingham, England, Cited by Fulford ( 1964).
References 194
Garrett-Price, B. A, 1 985, Fouling of Heat Exchangers, Characteristics, Costs,

Prevention, Control, Removal. Noyes Publications, New Jersey, Cited by Bott
( 1 995).
Gimbutis, G. J., Drobacivius, A. 1. & Sinkunas, S . S., 1978, Heat transfer of a

turbulent water film at different initial flow conditions and high temperature
gradients, Proc. 6th Int. Heat Transfer Con! , Vol. 1 , pp. 321-326, Toronto,
Canada,
Gordon, K. P., Hankinson, D. 1. & Carver, C. E., 1 968, Deposition of milk solids on
heated surfaces, Journal of Dairy Science, 51(4), 520-526.
Gotham, S. M., 1 990, Mechanisms of Protein Fouling of Heat Exchangers, Ph.D.

Dissertation, University of Cambridge, UK, Cited by Fryer et al. ( 1995).
Gotham, S. M., Fryer, P. J. & Pritchard, A M., 1 989, Model studies of food fouling,
Fouling and Cleaning in Food Processing, (Edited by Kessler, H. G. &
Lund, D. B.), pp. 1 - 13,. Prien, Bavarie, Germany.
Gotham, S. M., Fryer, P. J. & Pritchard, A M., 1 992, B-Iactoglobulin denaturation

and aggregation reactions and fouling deposit formation: a DSC study,
International Journal of Food Science and Technology, 27, 3 1 3-327.
Grandison, A S., 1 988a, Ultra-high-temperature processing of milk: seasonal

variation in deposit formation in heat exchangers, Journal of the Society of
Dairy Technology, 41, 43-49.
Grandison, A S., 1 988b, Effects of natural (or seaso?al) variation in concentration of

components of milk and addition of divalent cations on ultra-high-temperature
References 1 95
processing characteristics, Journal of the Society of Dairy Technology, 41,

1 17- 1 19.
Grandison, A. S., 1 988c, Fouling in Process Plant, (Edited by Pritchard, A. M.), pp.
1 97-2 14, Institute of Corrosion, Science & Technology, London, Cited by
Fryer et al. ( 1 995).
Gray, R. M., 198 1 , Technology of skim milk evaporation, Journal of the Society of
Dairy Technology, 34(2), 53-57.
Green, M. L., Foster, C. L. & Britten, M., 1988, Formation and adhesion of deposit
to heated surfaces in contact with milk, Fouling in Processing Plant, (Edited
by Pritchard, A. M.), pp. 1 83- 196, Institute of Corrosion, Science and
Technology, London, Cited by Fryer et al. ( 1 995).
Gr�nlund, M., 1984, Optimising evaporator and spray dryer performance in food
processing, Profitability of Food Processing, pp. 397-408, Pergamon press,
EFCE Publication Series No.36, The Institution of Chemical Engineers.
Gynning, K., Thome, K. E. & Samuelsson, E. G., 1958, Das Anbrennen in

Plattenerhitzem, Milchwissenschaft, 13, 62-70, Cited by Fryer et al. ( 1995).
Hahn, G., 1 985, Evaporator design, Concentration and Drying of Food, (Edited by.
MacCarthy, D.), pp. 1 13-1 3 1 , University College, Cork, Republic of Ireland.
Hall, C. W. & Hedrick, T. 1., Drying of Milk and Milk Products, AVI Publishing
Company Inc., Westport Connecticut, Cited by Mackereth (1993).
Hallstrom, B., 1988a, Heat transfer and liquid foods, Heat Transfer and Food
Products, (Edited by Hallstrom, B., Skjoldebrand, C. & Tragardh, C.), pp.
1 40- 143, London and New York.

References 196
Hallstrom, B., 1 988b, Preconcentration: new developments, Preconcentration and

Drying of Food Materials, (Edited by Bruin, S.), pp. 37-49, Elsevier Science
Publishers, Amsterdam, The Netherlands.
Hallstrom, B., 1 985, Heat exchange, Evaporation, Membrane Filtration and Spray
Drying in Milk Powder Cheese Production, (Edited by Hansen, R.), pp. 25-
43, North European Dairy Journal, Denmark.
Hallstrom, B., 1977, Evaporators with rotating conical heating surfaces,

Verfahrenstechnik, 11(8), 448, 450, 452, 454.
Hallstrom, B., 1 969, Chapter 1 The use of Centri-therm, expanding-flow and forced
circulation plant evaporators, The Use of Centri-Therm, Expanding-flow and
Forced-circulation Plant Evaporators in the Food and Biochemical Industries,
pp. 3- 1 3, United Nations Publication, Vienna and New York.
Hege, w. U. & Kessler, H. G., 1 986a, Deposit formation of protein containing dairy
liquids, MilchwissenschaJt, 41(6), 356-359.
Hege, w. U. & Kessler, H. G., 1 986b, Die Ansatzbildung beim Erhitzen von Milch
und Molke, Int. Z. Lebensmittel-Technol. Verfahrenstechnik, 37, 92-1 00, Cited
by Fryer et al. (1995).
Hegg, P. O., Castberg, H. B. & Lundh, G., 1 985, Fouling of whey proteins on
stainless steel at different temperatures, Journal of Dairy Research, 52, 2 1 3-
2 1 8.
Hegg, P. O. & Larsson, K., 1 98 1 , Ellipsometry studies of adsorbed lipids and milk
proteins on metal surfaces, Fundamentals and Application of Surface
Phenomena Associated With Fouling and Cleaning in Food Processing,
References 1 97
(Edited by Hallstrom, B., Lund, D. B . & Tragardh, C.), pp. 250-255,

Reprocentralen, Lund.
Holden, K. M., Wanniarachchi, A. S., Marto, P. J., Boone, P. J. & Rose, J. W., 1987,
The use of organic coatings to promote dropwise condensation of steam,
Journal of Heat Transfer, 109, 768-744.
Honig, P., 1 963, Principles of Sugar Technology, Elsevier Science Publishers,

Amsterdam, The Netherlands.
Hiddink, J., Lalande, M. & Streuper, A., 1 986, Heat treatment of whipping cream. 1.
Fouling of the pasteurization equipment, MilchwissenschaJt, 41(9), 542-546.
Housova, J., 1 970, Prestup tepla pri varn v odparkach se stekajicfm filmem, Prumysl
Polravin, 21( 1 ), 7- 12, Cited by Mackereth (1995).
Ito, R. & N akanishi, T., 1967, Fonnation of milk deposits on heat exchange surfaces
in UHT pasteurising plants, Japanese Journal of Dairy Science, 15A, 78-88,
Cited by Fryer et al. (1995).
Jachuck, R. J. J. & Ramshaw, c., 1 994, Process intensification: heat transfer

characteristics of tailored rotating surfaces, Heat Recovery System & CHP,
14(5), 475-491 .
Jackson, M. L., 1955, AIChEJ (Am. Inst. Chem. Eng. 1.), 1, 23 1 , Cited by Fulford
( 1964).
Jakob, M., 1 936, Mech. Eng. , 58, 729, Cited by Carey ( 1 992).
Jebson, R. S., 1 989, Chapter 2 The performances of falling film evaporator,

Evaporation and Spray Drying in the New Zealand Dairy Industry, (Edited by
References 1 98
Jebson, R. S .), pp. 14-32, Food Technology Department, Massey University,

Palmerston North, New Zealand.
Jebson, R. S., 1 988, The performances of falling film evaporators, Proceeding of

IPENZ ( The Institution of Professional Engineers of New Zealand)
Conference, Palmerston North, 2, 1 95-204.
Jebson, R. S., Huston, T., Janata, W. & Murray, G., 1 993, Proposal for the
Utilisation of LPG As a Source of Energy on Dairy Farms in General and
Specifically for the on Farm Site Dewatering of Milk, Energy Transfer
Developments Group, Massey University, Palmerston North.
Jebson, R. S. & Iyer, M., 1 99 1 , Performances of falling film evaporator, Journal of

Dairy Research, 58, 29-38.
Jelen, P., 1 98 1 , Experience with direct and indirect UHT processing of milk-a
Canadian viewpoint, Presented at 68th Annual meeting of IAMFES, Spokane,
WA, USA, Cited by Swartzel ( 1 983).
Jelen, P. & Rattray, W., 1 995, Thermal denaturation of whey proteins, Heat-Induced
Changes in Milk, (Edited by Fox, P. F.), 2nd Edition, pp. 66-85, International
Dairy Federation, Belgeiun.
Jeurnink, T. 1. M., 1 99 1 , Effect of proteolysis in milk on fouling in heat exchangers,

Netherland Milk and Dairy Journal, 45, 23-32.
Jeurnink, T. 1. M., 1 995a, Fouling of heat exchangers by fresh and reconstituted milk
and the influence of air bubbles, Milchwissenschaft, 50(4), 1 89- 1 92 .
Incropera, F. P . & DeWitt, D. P . , 1 990, Fundamentals of Heat Transfer and Mass

Transfer, (3rd Edition), John Wiley & Sons, New Yark.
References 199
Jeumink., T. 1. M., 1 995b, Relation between the denaturation and fouling behaviour
of B-Iactoglobulin, Heat Treatments and Alternative Methods, p. 1 .7,
International Dairy Federation, Vienna, Austria.
Jeurnink., T. 1. M. & Brinkman, D. W., 1994, The cleaning of heat exchangers and
evaporators after processing milk or whey, International Dairy Journal, 4,
347-368.
Jeurnink, T. J. M. & de Kruif, C. G., 1 995, Calcium concentration in milk in relation

to heat stability and fouling, Netherland Milk and Dairy Journal, 49, 1 5 1 -165.
Jeumink, T. 1. M., Walstra, P. & de Kruif, C. G., 1996, Mechanism of fouling in

dairy processing, Netherland Milk and Dairy Journal, 50, 407-426.
Johnson, 1. J. & Roland, C. T., 1940a, Study of dairy cleaning problems. I. Films and
deposits on hot milk equipment, Journal of Dairy Science, 23, 457-46 1 .
Johnson, 1 . 1 . & Roland, C . T., 1940b, Study of dairy cleaning problems. II.
Effectiveness of alkalies in removing heat-deposited milk solids and butterfat
films, Journal of Dairy Science, 23, 463-469.
Kapitsa, P. L., 1948, Zh. Eksperim. i Teor. Fiz., 1 83, Cited by Fulford ( 1 965).
Kast, W., 1963, Heat transfer during dropwise condensation (in Gennan), Chem. -Ing.
Tech., 35, 1 63 - 1 68, Cited by Carey ( 1992).
Kastanas, P., Jewis, M. J. & Grandison, A. S., 1 995, Design and development of a
miniature UHT plant for fouling studies and its evaluation using milk adjusted
to different pH values, goat's milk and milk permeate, Trans IChemE, 73,
Part C, 67-75 .
References 200
Kern, D. Q. & Karakus, H. J., 1 958, Mechanically aided heat transfer, AIChEJ. (Am.
Inst. Chern. Eng.), Preprint No. 1 7, New York, Cited by Reay ( 1 99 1).
Kessler, H. G., 1 989, Multistage evaporation and water vapour recompression with
special emphasis on high dry matter content, product losses, cleaning and
energy savings, Dairy Industries, 545-558.
Kessler, H. G., 1985, Energy aspects of food preconcentration, Concentration and

Drying of Food, (Edited by MacCarthy, D.), pp. 147- 163, Elsevier Applied
Science, London and New York.
Kessler, H. G., Fiedler, J. & Hege, W. u., 1 986, Formation of deposits from liquid
foods on heated surfaces during heating and evaporation in UHT pasteurising
plants, Chern. Ing. Tech., 58, 475-485, Cited by Fryer et al. ( 1 995).
Kessler, H. G. & Lund, D. B., 1 989, Fouling and Cleaning in Food Processing,
University of Munich, Prien, Bavaria, Germany.
Kreith, F., 1968, Convection heat transfer in rotating systems, Advances in Heat
Transfer, 5, 129-25 1 .
Kroll, J. F & McCutchan, J. W., 1 968, Heat transfer in an LTV falling film
evaporator: A theoretical and experimental analysis, Journal of Heat Transfer,
Transactions of the ASME, 58 (3), 201 -210.
Kuo, C. Y., Lida, H. T., Tayler, J. H. & Kreith, F, 1 960, Journal of Heat Transfer,
82, 1 39, Cited by Kreith ( 1968).
Lalande, M. & Corrieu, G., 1 98 1 , Fouling of a plate heat exchanger by milk,

Fundamentals and Applications of Surface Phenomena Associated With
Fouling and Cleaning in Food Processing, (Edited by Hallstrom, B.,

References 201
Tragardh, C. & Lund, D. B .), pp. 279-288, University of Lund, Sweden, Cited
by Fryer et al. (1995).
Lalande, M., Tissier, J. P. & Corrieu, G., 1985, Fouling of heat transfer surfaces
related to B-Iactoglobulin denaturation during heat processing of milk,
Biotechnology Progress, 1, 1 3 1- 1 39, Cited by Fryer et al. ( 1995).
Lalande, M., Tissier, J. P. & Corrieu, G., 1984, Fouling of a plate heat exchanger
used in ultra-high-temperature sterilization of milk, Journal of Dairy
Research, 51, 557-568.
Langton, M. & Hermansson, A., 1992, Fine-stranded and particulate gels of B

lactoglobulin and whey protein at varying pH, Food Hydrocolloids, 5(6), 523-
539, Cited by N�rb�ge ( 1994).
Leniger, H. A., 1 986, Concentration by evaporation, Preconcentrtion and Drying of

Food Materials, (Edited by Bruin, S.), pp 54-76, Elsevier Science Publishers
B.V., Amsterdam, The .Netherlands.
Levich, V. G., 1 962, Physiochemical Hydrodynamics, Prentice-Hall, Englewood

Cliffs. NJ., USA.
Lewicki, P. P. & Kowalczyk, R., 1 979, Heat transfer in the Centri-Therm CT- 1 B
evaporator, International Union of Food Science and Technology, Food
Engineering Symposium, Warsaw Agriculture University, Warsaw, Poland.
Lim, S. T., 1 980, Hydrodynamics and Mass Transfer Processes Associated With the
Absorption of Oxygen in Liquid Films Flowing Across a Rotating Disc, Ph.D.
Dissertation, University of Newcastle, Upon Tyne, U.K, Cited by Jachuck and

Ramshaw ( 1994).
References 202
Lyle, 0., 1 947, The Efficient Use of Steam, His Majesty's Stationary Office, London.
276-286, Cited by Macl(ereth ( 1 993).
Lyster, R. L. J., 1 970, The denaturation of a-lactalbumin and �-lactoglobulin in

heated milk, Journal of Dairy Research, 37, 233-243.
Lyster, R. L. J., 1 965, The composition of milk deposits in an ultra-high-temperature

plant, Journal of Dairy Research, 32, 203-208.
Maas, R., Lalande, M. & Hiddink, J., 1 985, Fouling of a plate heat exchanger by
whipping cream, Fouling and Cleaning in Food Processing, (Edited by
Lund, D. B., Plett, E. A. & Sandu, C.), pp. 217-225, University of Wisconsin,
Cited by Fryer et al. (1995).
Mackereth, A. R., 1995, Chapter 9 Evaporation, Milk Powder Technology Principles

& Process Applications, (Edited by Tuoc Trinh), New Zealand Dairy
Research Institute, Palmerston North,
Mackereth, A. R., 1993, Thennal and Hydraulic Aspects of Falling Film

Evaporation, Ph.D. Thesis, University of Canterbury, Christchurch, New
Zealand.
Mai, K. L., 1 958, Chem.Egn.Progr.Symp, 54, 57, Cited by Kreith ( 1 968).
MaIkki , Y. & Veldstra, 1., 1967, Flavour retention and heat transfer during
concentration of liquid in a centritherm film evaporator, Food Technology, 21,
1 179- 1 1 84.
Mannheim, C. H. & Passy, N., 1 974, Non-membrane concentration, Advances in

Preconcentration and Dehydration of Foods, (Edited by Spicer, A.), pp. 1 52-
1 93, Applied Science Publishers Ltd, London, UK.

References 203
Marrs, W. M. & Lewis, D. F., 1986, Fouling of Heated Surfaces by Proteinaceous

Fluids, No.569, The British Food Manufacturing Industries Research
Association, London, UK.
Martynov, Y. V., 1 984, Flow of a Liquid film over the inner surface of a rotating
cylinder, Journal of Applied Mechanics Technical Physics, 23(3), 400-407.
McCabe, W. L. & Smith, 1. c., 1965, Unit Operations of Chemical Engineering.

(2nd Edition), Mcgraw-Hill Book Company, New York and London. New
York and London.
Melo, L. F., Bott, T. R. & Bernardo, C. A., 1 988, Fouling Science and Technology,
Kluwer Academic Publishers, Dordrecht, The Netherlands.
Mincowycz, W. 1. & Sparrow, E. M., 1 966, Condensation heat transfer in the

presence of non-condensables, interfacial resistance, superheating, variable
properties, and diffusion, International Journal of Heat and Mass Transfer, 9,
1 125- 1 144.
Minton, P. E., 1 986, Handbook of Evaporation Technology, Noyes Pulications, Park

Ridge, New jersey, USA
Moresi, M., 1 985, Design and optimisation of falling-film evaporators, Developments

in Food Preservation-3, (Edited by Thorne, S.), pp. 1 83-244, Elsevier Applied
Science Publishers, London and New York.
Muir, D. D. & Sweetsur, A. W. M., 1 976, The influence of naturally occurring levels
of urea on the heat stability of bulk milk, Journal of Dairy Research, 43, 495-
499.
References 204
MUller-Steinhagen, H., 1 989, Heat Transfer and Heat Exchanger Design, Continuing
Education Seminar Notes, Chemical and Materials Engineering Department,
University of Auckland, New Zealand, Cited by Machereth ( 1 995).
Mulvihill, D. M. & Donovan, M., 1987, Whey proteins and their thermal
denaturation. A review., Irish Journal Food Science Technology, 11, 43-75.
Nikolaev, V. S., Vachagin, K. D. & Baryshev, Y. N., 1967, Film flow of viscous
liquids over surfaces of rapidly rotating conical discs, International Chemical
Engineering, 7(4), 595-598.
Nikuradse, J., 1 933, Forschungsheft, 361, 1 , Cited by Dukler and Bergelin ( 1 952).
Nocol, A. A. & Gacesa, M., 1970, Condensation of steam on a rotating vertical

cylinder, Journal of Heat Transfer, 92, 144- 152.
Nong, T. C, 1 969, Factors affecting the degree of concentration of apple juice in a

centrifugal evaporator, Przemysl-Spozywczy, 23(10), 438-440, (in Polish).
NS?)rbS?)ge, L., 1 994, Whey Protein Gelation and Denaturation at Different pH and
Calcium Chloride Concentrations, Masterate Thesis, Massey University,
Palmerston North, New Zealand.
Nukiyama, S., 1 934, The maximum and minimum values of heat Q transmitted from
'
metal to boiling water under atmospheric pressure, Japan Soc. Mech. Eng. ,
37, 367-374, English translation in International Journal of Heat and Mass
Transfer, 9, 1 4 1 9- 1433, (1966).
Nusselt, W., 1 9 1 6, Die Oberflachenkondensation des Wasserdampfes, Ver. Deut.

Ingr. 2., 60, 54 1 -546, 569-575, Cited by Baumann ( 1 990).
References 205
O' Connor, G. E. & Russell, T. W. F., 1 978, Heat transfer in tubular fluid-fluid
systems, Advances in Chemical Engineering, (Edited by Drew, T. B . et al.),
10, pp. 1-53, Academic Press, New York and London.
Oliver-Daumen, B., 1 986, Optimizing the operating parameters for alcohol reduction
with the Centri-therm, Monatsschrift-Fuer-Brauwissenschajt, 39(4), 1 32- 142.
(in German).
Oliver-Daumen, B., 1 982, Practical experience with use of Centri-Therm equipment

for reduction of the alcohol content of beer, Monatsschrift-Fuer
Brauwissenschajt, 35(3), 1 0 1 - 1 06, (in German).
O'Neill, G. A. & Westwater, J. W., 1 984, Dropwise condensation of steam on

electroplated silver surfaces, International Journal of Heat and Mass
Transfer, 27, 1 539- 1 549.
Pajunen, E., Vanhatalo, I., Linko, M. & Pessa, E., 1 974, Concentration of yeast
extract with a centrifugal film evaporator, Proc. Int. Congress Food Science
and Technology, 4, 390-396.
Parker, M. E. & Johnson, A. H., 1930, Cleaning of deposits on hot milk equipment,
Proceeding of the International Association of Milk Dealers 23rd Annual
Convention, Cited by Fryer et al. (1995).
Pattenden, R. F., 1964, Journal of Mechanic Engineering and Science, 6, 144, Cited
by Kreith (1968).
Peters, T. Jr., 1985, Serum albumin, Advances in Protein Chemistry, 37, 1 6 1 -245.
References 206
Pritchard, A M., 1988, The economics of fouling, Fouling Science and Technology,
(Edited by Melo, L. F., Bott, T. R & Bernardo, C. A), pp. 3 1 -45, Kluwer
Academic Publishers, Dordrecht, The Netherlands.
Pritchard, A. M., 1979, Heater exchanger fouling in British industry, Fouling

prevention research digest, Vol. 1 , pp. 4-6, Cited by Pritchard ( 1 988).
Reay, D. A, 1 99 1 , Heat transfer enhancement-A review of techniques and their

possible impact on energy efficiency in the UK, Heat Recovery Systems &
CHP, 11(1), 1-40.
Reed, R G., Feldhoff, R c., Clute, O. L. & Peters, T. Jr., 1975, Fragments of
bovine serum albumin produced by limited proteolysis. Conformation and
ligand binding, Biochemistry, 14 , 4578-4583.
Reed, R G., Putnam, F. W. & Peters, T. Jr., 1980, Sequence of residues 400-403 of
bovine serum albumin, Biochemical Journal, 191, 867-868.
Robbins, R M. & Holmes, L. G., 1 970, Circular dichroism spectra of a-lactalbumin,

Archives of Biochemistry and Biophysics, 221, 234, Cited by N!1Srb!1Sge ( 1 994).
Robertson, N. M. & Dixon, A., 1 969, The nitrogen fractions and the heat stability of
bovine milk, Agroanimalia, 1, 141-144, Cited by Fryer et al. ( 1995).
Robertson, P. S., 1 987, Developments in dairy processing, Milk-The Vital Force,

Proceeding of the XXII International Dairy Congress, The Hague,
Supplement.
Rogers, G. F. C. & Mayhew, Y. R, 1 984, Thermodynamic and Transport Properties

of Fliuds - SI Units, 3rd edtion, Basil Blackwell, Oxford, London.
References 207
Roscoe, S . G. & Fuller K. L., 1 994, Fouling of model surface: adsorption and
removal of whole unpasteurized milk, Food Research International, 27, 363-
369.
S andu, C. & Lund, D. B., 1982, Fouling of heat treatment equipment by food fluids:
computational models, Food Processing Engineering, (Edited by
Schwartzberg, H. G., Lund, D. B . & Bomben, 1. L.), 78, pp. 12-30, AIChE
Symposium Series, No. 2 1 8, American Institute of Chemical Engineer, New
York.
Sandu, G., 1 989, Chemical reaction fouling due to milk: defects-growth model,
Fouling and Cleaning in Food Processing, (Edited by Kessler, H. G., &
Lund, D. B.), Prien, Baravia, Germany.
Sato, T. & Matsumura, H., 196 1 , Bull, Japan. Soc. Mech. Eng., 7, 392, Cited by
Carey ( 1 992).
Schlichting, H., 1968, Boundary Layer Theory, McGrew-Hill Book Company, New
York.
Schwartzberg, H. G., 1 988, Food property effects in evaporation, Food Properties
and Computer-Aided Engineering of Food Processing Systems, (Edited by
Singh, R. P. & Medina, A. G.), pp. 443-470, Kluwer Academic Publishers,
Dordrecht, The Netherlands.
Semenov, P. A., 1 944, Zhurn. Tekh. Fiz., 14, 7-8, Cited by Zhivaikin ( 1962).
Shinn, B. E., 1 97 1 , The Centri-therm evaporator and its application to heat sensitive
foods, Journal of Applied Chemical Biotechnology, 21( 1 2), 366-37 1 .

References
208
Silver, R S., 1 964, An approach to a general theory of surface condensers, Proc.
Inst. Mech. Eng., 178, 339-376, Cited by Carey ( 1 992).
Sinek, J. R & Young, E. H., 1960, Heat transfer in falling film long tube vertical
evaporators, Chemical Engineering Progress, 58(12), 74-80.
Skudder, P. J., Brooker, B . E., Bonsey, A. D. & Alvarez-Guerrero, N. R., 1 986,
Effect of pH on the formation of deposit from milk on heated surfaces during
UHT processing, Journal of Dairy Research, 53, 75-87.
Skudder, P. J., Thomas, E. L., Pavey, J. A. & Perkin, A. G., 1 98 1 , Effect of adding
potassium iodate to milk before UHT treatment, Journal of Dairy Research,
48, 99- 1 1 3 .
Sparrow, E. M . , Minkowycz, W. J . & Saddy, M., 1 967, Forced convection
condensation in the presence of non-condensables and interfacial resistance,
International Journal of Heat and Mass Transfer, 10, 1 829- 1 845.
Sparrow, E. M. & Hartnett, 1. P., 1 96 1 , Condensation on a rotating cone, Journal of
Heat Transfer, Transactions of the ASME, Series C, 2, 1 0 1 - 1 02.
Sparrow, E. M. & Gregg, J. L., 1 959, A theory of rotating condensation, Journal of
Heat Transfer, 81, 1 13.
Steinhagen, R , Mller-Steinhagen, H. & Maani, K., 1 990, Heat Exchanger
Applications, Fouling Problems and Fouling Costs in New Zealand Industries,
RD8829, Published by Market Information and Analysis Group, Energy and

Resources Division, Ministry of Commerce, Wellington, New Zealand,
Stephan, K., 1 992, Heat Transfer in Condensation and Boiling, Springer-Verlag,
Berlin.
References 209
Sugawara, S . & Katusuta, K., 1 966, Fundamental study on dropwise condensation,
Proc. 3rd Int. Heat Transfer Conj, 2, 354-361 , Cited by Carey ( 1992).
Swartzel, K. R., 1 983, Tubular heat exchanger fouling by milk during ultra-high
temperature process, Journal of Food Science, 48, 1507- 15 1 1 .
Taborek, J., Aoki, T., Ritter, R. B . , Palen, J. W. & Knudsen, 1. G., 1 972, Fouling:
The major unresolved problem in heat transfer, Chemical Engineering
Progress, 68 (2) , 59-67.
Tanaka, H., 198 1 , Effect of Knudsen number on dropwise condensation, Journal of
Heat Transfer, 103, 606-607.
Tanaka, H., 1979, Further developments of dropwise condensation theory, Journal of
Heat Transfer, 101, 603-6 1 1 .
Tanaka, H., 1975, A theoretical study of dropwise condensation, Journal of Heat
Transfer, 97, 72-78.
Taylor, D. c., 1982, The history of evaporation, 1 750 to 1 900, The Chemical
Engineer, S, 1 87- 1 90.
Thome, J. R., 1990, Enhanced Boiling Heat Transfer, Hemisphere Publishing
Corporation, New York, Washington, Philadelphia and London.
Timasheff, S. N., Susi, H. & Stevens, L., 1967, Infrared spectra and protein
conformation in aqueous solutions, Journal of Biological Chemistry, 242,
5467-5473, Cited by Jelen & Rattray ( 1 995).

--- ----
- .. �---
References 210
Timasheff, S. N., Townend, R. & Meschanti, L . , 1 966, The optical rotary dispersion
of the B-Iactoglobulin, Journal of Biological Chemistry, 241, 1 863- 1 870, Cited
by Jelen & Rattray ( 1 995).
Tissier, 1. P. & Lalande, M., 1 986, Experimental device and methods for studying
milk deposit formation on heat exchange surfaces, Biotechnology Progress,
2(4), 21 8-229.
Tissier, J. P., Lalande, M. & Corrieu, G., 1 984, A study of milk deposit on a heat
exchange surface during ultra-high-temperature treatment, Engineering and
Food. Vol I. Engineering Sciences in the Food Industry, (Edited by
McKenna, B. M.), pp. 59-68, Applied Science Publishers, London, Cited by
Fryer et al. ( 1 995).
Traegardh, C., 1974, Production of leaf protein concentrate for human consumption
by isopropanol treatment. A comparison between untreated raw juice and raw
juice concentrated by evaporation and ultra-filtration, Lebensmittel
Wissenschajt-Technoloie; 7(4), 199-201 , (in Swedish).
Veda, T. & Tanaka, H., 1 975, Measurements of velocity, temperature and velocity
fluctuation distributions in falling liquid films, International Journal of
Multiphase Flow, 2, 261 -272.
Vmur, A. & Griffith, P., 1 965, Mechanism of dropwise condensation, Journal of
Heat Transfer, 87, 275-282.
Van Stralen, S. J. D. & Cole, R., 1979, Boiling Phenomena, Vois. 1 and 2
Hemisphere, Washington.
References 21 1
Vasiliev, L. L. & Khrolenok.V.V., 1 993, Heat transfer enhancement with
condensation by surface rotation, Heat Recovery System & CHP, 13(6), 547-
563.
Visser, J., 1988, Adhesion and removal of particles I, Fouling and Cleaning in
Process Plant, pp. 83- 1 00, St. Catherine's college, Oxford, United Kingdom.
Visser, J., Lindell, A & Paulsson, M. A , 1 995, Heat induced depositions of milk
components onto a rotating disk, Heat Treatments and Alternative Methods, p.
1 .8, International Dairy Conference, Vienna, Austria.
Visser, J., Minihan, A, Smits, P., Tjan, S. B . & Heertje, I., 1 986, Effects of pH and
temperature on the milk salt system, Netherland Milk and Dairy Journal, 40,
3 5 1 -368.
Walstra, P. & Jenness, R., 1 984, Dairy Chemistry and Physics, John Wiley & Sons
Inc., New York.
Welch, J. F. & Westwater, J. W., 1 96 1 , Microscopic study of dropwise condensation,
International Developments in Heat Transfer. Proc. Int. Heat Transfer Con!
ASME., Part II, 302-309, Cited by Carey ( 1 992).
Westergard, V., 1983, Milk Powder Technology Evaporation and Spray Drying (3rd
and Revised Edition), A/S NIRO Atomizer, Copenhagen, Denmark.
Whalley, P. B., 1 992, Fouling -still the major unresolved problem in heat transfer,
Trans IChemE, 70, Part A, 545.
Wiegand, B., 1985, Chapter 6 Evaporation, Evaporation, Membrane Filtration and
Spray Drying in Milk Powder Cheese Production, (Edited by Hansen, R.), pp.
9 1 - 178, North European Dairy Journal, Denmark.

References 212
Wiegand, B . , 1 978, Multiple effect evaporation and the DRS tube and the noise
problem., Lecture Handout in November, 1977 at various places in New
Zealand.
Wilke, W., 1 962, Warmetibergang an reiselfilme, Ver. Dent.lngr. Forschungsh, 490,
Cited by Chun & Seban ( 197 1).
Wood, P. W., 1 982, Physical Properties of Dairy Products, Technical Report, ISSN
0 1 1 0-8948, Ministry of Agriculture and Fisheries, Hamilton, New Zealand,
Xiong, Y. L., 1 992, Influence of pH and ionic environment on thermal aggregation
of whey proteins, Journal of Agricultural and Food Chemistry, 40, 380-384.
Xu, X., 1 996, The Investigation of the Quality of Milk Pre-Concentrated by an On
Farm Evaporation System, Masterate Thesis, Massey University, Palmerston
North, New Zealand.
Yanniotis, S. & Kolokotsa, D., 1 996, Boiling on the surface of a rotating dise,
Journal of Food Engineering, 30, 3 13-325.
Yanniotis, S., 1 983, Thermal Characteristics of Absorption-Driven and Steam-Heated
Long Tube Vertical Falling Film Evaporators, Ph.D. Dissertation, University
of Massachusetts, Amberst, USA, Cited by Schwartzberg ( 1 988).
Yoon, J. & Lund, D. B . , 1989, Effect of operating conditions, surface coatings and
pretreatment on milk fouling in a plate heat exchanger, Fouling and Cleaning
in Food Processing, (Edited by Kessler, H. G. & Lund, D. B.), pp. 59-80,
University of Munich, Germany.
Zhivaikin, L. Ya., 1 962, Liquid film thickness in film-type units, International
Chemical Engineering, 2(3), 337-34 1 .

References 213
Zhivaikin, L . Ya. & Volgin, B . V., 1 964, Hydraulic resistance in descending two
phase flow in film-type equipment, International Chemical Engineering, 4( 1 ) ,
80-84.
------
APPENDICES
--- ---- -- ----
Appendix I
Experimental results of heat transfer in evaporators
2
2 1 The resu1ts 0f 4 expenment In the Centn· therm eva )orator WI' th water
TRIALS ST EV t.T FF RS W exp Heal Ave-Heal
(Times) (OC ) (OC ) (K) 3 3
(x1 0· m /s) (radls) (xl O·3m3/s) (kW/m2 K) (kW/m2 K)
1 -1 70 60 10 0.01 67 1 04.720 0.0030 7.86 8.01
1 -2 70 60 10 0.01 67 1 04.720 0.0031 8.08
1 -3 70 60 10 0.01 67 1 04.720 0.0031 8.08
2-1 70 60 10 0.0500 1 04.720 0.0031 8.1 9 8. 1 1
2-2 70 60 10 0.0500 1 04.720 0.0031 8. 1 9
2-3 70 60 10 0.0500 1 04.720 0.0030 7.97
3-1 80 70 10 0.01 67 1 04.720 0.0034 8.86 8.79
3-2 80 70 10 0.01 67 1 04.720 0.0035 9.08
3-3 80 70 10 0.01 67 1 04.720 0.0033 8.43
4-1 80 70 10 0.0500 1 04.720 0.0034 8.86 8.68
4-2 80 70 10 0.0500 1 04.720 0.0034 8.75
4-3 80 70 10 0.0500 1 04.720 0.0033 8.43
5-1 70 60 10 0.01 67 1 86.401 0.0031 8.08 8.08
5-2 70 60 10 0.01 67 1 86.401 0.0032 8.30
5-3 70 60 10 0.01 67 1 86.401 0.0030 7.86
6-1 70 60 10 0.0500 1 86.401 0.0032 8.30 8.1 5
6-2 70 60 10 0.0500 1 86.401 0.0031 8.08
6-3 70 60 10 0.0500 1 86.401 0.0031 8.08
7-1 80 70 10 0.01 67 1 86.401 0.0038 9.73 9.54
7-2 80 70 10 0.01 67 1 86.401 0.0036 9.40
7-3 80 70 10 0.01 67 1 86.401 0.0037 9.51
8-1 80 70 io 0.0500 1 86.401 0.0036 9.29 9.33
8-2 80 70 10 0.0500 1 86.401 0.0035 9.08
8-3 80 70 10 0.0500 1 86.401 0.0037 9.62
9-1 80 60 20 0.01 67 1 04.720 0.0066 8.65 8.52
9-2 80 60 20 0.01 67 1 04.720 0.0064 8.41
9-3 80 60 20 0.01 67 1 04.720 0.0065 8.52
1 0-1 80 60 20 0.0500 1 04.720 0.0064 8.41 8.44
1 0-2 80 60 20 0.0500 1 04.720 0.0065 8.52
1 0-3 80 60 20 0.0500 1 04.720 0.0064 8.41
1 1 -1 90 70 20 0.01 67 1 04.720 0.0073 9.40 9.35
1 1 -2 90 70 20 0.01 67 1 04.720 0.0073 9.40
1 1 -3 90 70 20 0.01 67 1 04.720 0.0071 9.24
1 2-1 90 70 20 0.0500 1 04.720 0.0074 9.S6 9.S3
1 2-2 90 70 20 O.OSOO 1 04.720 0.0074 9.62
1 2-3 90 70 20 0.0500 1 04.720 0.0073 9.42
1 3-1 80 60 20 0.01 67 1 86.401 0.0068 8.95 8.84
1 3-2 80 60 20 0.01 67 1 86.401 0.0068 8.84
1 3-3 80 60 20 0.01 67 1 86.401 0.0067 8.73
1 4-1 80 60 20 0.0500 1 86.401 0.0069 9.06 8.81
1 4-2 80 60 20 0.0500 1 86.401 0.0067 8.73
1 4-3 80 60 20 0.0500 1 86.401 0.0066 8.62
1 S-1 90 70 20 0.01 67 1 86.401 0.0076 9.85 9.80
Appendix I 215
1 5-2 90 70 20 0.01 67 1 86.401 0.0075 9.73

1 5-3 90 70 20 0.01 67 1 86.401 0.0076 9.83
1 6-1 90 70 20 0.0500 1 86.401 0.0076 9.90 9.86
1 6-2 90 70 20 0.0500 1 86.401 0.0076 9.83
1 6-3 90 70 20 0.0500 1 86.401 0.0076 9.85
1 7-1 80 65 15 0.0333 1 46.608 0.0048 8.40 8.33
1 7-2 80 65 15 0.0333 1 46.608 0.0048 8.25
1 7-3 80 65 15 0.0333 1 46.608 0.0048 8.33
1 8-1 80 65 15 0.0333 1 46.608 0.0048 8.33 8.46
1 8-2 80 65 15 0.0333 1 46.608 0.0049 8.47
1 8-3 80 .65 15 0.0333 1 46.608 0.0049 8.57
.
2 2 The expenmental resu ts 0btamed m
· the Centnt
· herm evaporator WI water
Hcal-�T
TRIALS ST EV �T FF RS W elCP Heal Ave-Heal
(OC ) (OC ) (K) • (x1 0·3 m%) (rad/s) (x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 65 60 5 0.0500 1 86.401 0.001 4 7.34 7.57
1 -2 65 60 5 0.0500 1 86.401 0.001 4 7.51
1 -3 65 60 5 0.0500 1 86.401 0.001 5 7.86
2-1 70 60 10 0.0500 1 86.401 0.0032 8.41 8.1 7
2-2 70 60 10 0.0500 1 86.401 0.0031 8.1 2
2-3 70 60 10 0.0500 1 86.401 0.0030 7.97
3-1 75 60 15 0.0500 1 86.401 0.0048 8.44 8.59
3-2 75 60 15 0.0500 1 86.401 0.0049 8.52
3-3 75 60 15 0.0500 1 86.401 0.0050 8.81
4-1 80 60 20 0.0500 1 86.401 0.0068 8.84 8.81
4-2 80 60 20 0.0500 1 86.401 0.0065 8.57
4-3 80 60 20 0.0500 1 86.401 0.0069 9.01
5-1 85 60 25 0.0500 1 86.401 0.0088 9.21 9.1 1
5-2 85 60 25 0.0500 1 86.401 0.0086 9.04
5-3 85 60 25 0.0500 1 86.401 0.0087 9.08
6-1 90 60 30 0.0500 1 86.401 0.01 07 9.32 9.30
6-2 90 60 30 0.0500 1 86.401 0.01 08 9.39
6-3 90 60 30 0.0500 1 86.401 0.01 05 9.21
7-1 95 60 35 0.0500 1 86.401 0.01 23 9.23 9.21
7-2 95 60 35 0.0500 1 86.401 0.01 23 9. 1 7
7-3 95 60 35 0.0500 1 86.401 0.01 23 9.23
8-1 1 00 60 40 0.0500 1 86.401 0.01 41 9.22 9.1 3
8-2 1 00 60 40 0.0500 1 86.401 0.0 1 39 9.09
8-3 1 00 60 40 0.0500 1 86.401 0.01 39 9.09
9-1 1 05 60 45 0.0500 1 86.401 0.01 52 8.85 8.89
9-2 1 05 60 45 0.0500 1 86.401 0.01 53 8.88
9-3 1 05 60 45 0.0500 1 86.401 0.01 53 8.93
1 0-1 110 60 50 0.0500 1 86.401 0.01 54 8.08 8.1 7
1 0-2 110 60 50 0.0500 1 86.401 0.01 58 8.25
1 0-3 110 60 50 0.0500 1 86.401 0.01 56 8.1 9
Appendix I 216
Heal-R S
TRIALS ST EV L\T FF RS W exP Heal Ave-Heal
(Oe ) (Oe ) (K) (x1 0·3m 3/s) (rad/s) _(x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 80 70 10 0.0500 41 .888 0.0026 6.81 6.77
1 -2 80 70 10 0.0500 41 .888 0.0027 6.92
1 -3 80 70 10 0.0500 41 .888 0.0025 6.59
2-1 80 70 10 0.0500 73.304 0.0032 8.21 8.1 8
2-2 80 70 10 0.0500 73.304 0.0031 8. 1 0
2-3 80 70 10 0.0500 73.304 0.0032 8.21
3-1 80 70 10 0.0500 1 04.720 0.0034 8.86 8.68
3-2 80 70 10 0.0500 1 04.720 0.0034 8.75
3-3 80 70 10 0.0500 1 04.720 0.0033 8.43
4-1 80 70 10 0.0500 1 46.608 0.0036 9.29 9.04
4-2 80 70 10 0.0500 1 46.608 0.0034 8.75
4-3 80 70 10 0.0500 1 46.608 0.0035 9.08
5-1 80 70 10 0.0500 1 78.024 0.0035 9.08 9.33
5-2 80 70 10 0.0500 1 78.024 0.0037 9.51
5-3 80 70 10 0.0500 1 78.024 0.0036 9.40
Heal-FF
TRIALS ST EV L\T FF RS W exP Heal Ave-Heal
(Oe ) (Oe ) (K) (xi 0·3m 3/s) (rad/s) (xi 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 70 60 10 0.01 33 1 46.608 0.0034 8.95 8.95
1 -2 70 60 10 0.01 33 1 46.608 0.0035 9.06
1 -3 70 60 10 0.01 33 1 46.608 0.0034 8.84
2-1 70 60 10 0.0250 1 46.608 0.0034 8.95 8.94
2-2 70 60 10 0.0250 1 46.608 0.0034 9.02
2-3 70 60 10 0.0250 1 46.608 0.0034 8.84
3-1 70 60 10 0.0333 1 46.608 0.0035 9.06 9.08
3-2 70 60 10 0.0333 1 46.608 0.0034 9.00
3-3 70 60 10 0.0333 1 46.608 0.0035 9.17
4-1 70 60 10 0.041 7 1 46.608 0.0035 9.1 7 9.1 5
4-2 70 60 10 0.041 7 1 46.608 0.0035 9.21
4-3 70 60 10 0.041 7 1 46.608 0.0035 9.06
5-1 70 60 10 0.0500 1 46.608 0.0035 9.28 9. 1 7
5-2 70 60 10 0.0500 1 46.608 0.0035 9.06
5-3 70 60 10 0.0500 1 46.608 0.0035 9.17
6-1 70 60 10 0.06 1 7 1 46.608 0.0035 9.28 9.37
6-2 70 60 10 0.061 7 1 46.608 0.0036 9.50
6-3 70 60 10 0.06 1 7 1 46.608 0.0036 9.34
Heal-FF-RS
TRIALS ST EV llT FF RS W exP Heal Ave-Heal
(Oe ) (Oe ) (K) (x1 0·3m 3;s) (rad/s) (x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 80 70 10 0.01 67 52.360 0.0029 7.48 7.56
1 -2 80 70 10 0.01 67 52.360 0.0029 7.56
1 -3 80 70 10 0.01 67 52.360 0.0030 7.65
2-1 80 70 10 0.041 7 52.360 0.0029 7.48 7.58
2-2 80 70 10 0.041 7 52.360 0.0029 7.56
2-3 80 70 10 0.0417 52.360 0.0030 7.69
3-1 80 70 10 0.0667 52.360 0.0029 7.56 7.90
3-2 80 70 10 0.0667 52.360 0.0031 8.00
Appendix I 217
3-3 80 70 10 0.0667 52.360 0.0031 8.1 3

4-1 80 70 10 0.01 67 1 04.720 0.0034 8.86 9.08
4-2 80 70 10 0.01 67 1 04.720 0.0035 9.08
4-3 80 70 10 0.01 67 1 04.720 0.0036 9.29
5-1 80 70 10 0.0417 1 04.720 0.0034 8.77 9.00
5-2 80 70 10 0.04 1 7 1 04.720 0.0035 9.08
5-3 80 70 10 0.04 1 7 1 04.720 0.0035 9.1 6
6-1 80 70 10 0.0667 1 04.720 0.0035 9.08 9;29
6-2 80 70 10 0.0667 1 04.720 0.0036 9.29
6-3 80 70 10 0.0667 1 04.720 0.0037 9.51
7-1 80 70 10 0.01 67 1 57.080 0.0036 9.21 9.48
7-2 80 70 10 0.01 67 1 57.080 0.0037 9.51
7-3 80 70 10 0.01 67 1 57.080 0.0038 9.73
8-1 80 70 10 0.041 7 1 57.080 0.0035 9.08 9.35
8-2 80 70 10 0.04 1 7 1 57.080 0.0036 9.42
8-3 80 70 10 0.041 7 1 57.080 0.0037 9.55
9-1 80 70 10 0.0667 1 57.080 0.0036 9.29 9 .58
9-2 80 70 10 0.0667 1 57.080 0.0038 9.73
9-3 80 70 10 0.0667 1 57.080 0.0038 9.73
Hcal-FF-RS-�T
TRIALS ST EV �T FF RS W exp Heal Ave-Heal
(OC ) (OC ) (K) (x1 0·3m 3/s) (radls) (x1 0-3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 80 70 10 0.0067 1 86.401 0.0039 1 0. 1 6 1 0.09
1 -2 80 70 10 0.0067 1 86.401 0.0038 9.94
1 -3 80 70 10 0.0067 1 86.401 0.0039 1 0.1 6
2-1 80 70 10 0.01 00 1 86.401 0.0038 9.73 9.65
2-2 80 70 10 0.01 00 1 86.401 0.0037 9.51
2-3 80 70 10 0.01 00 1 86.401 0.0038 9.73
3-1 80 70 10 0.01 33 1 86.401 0.0037 9.51 9.57
3-2 80 70 10 0.01 33 1 86.401 0.0037 9.62
3-3 80 70 10 0.01 33 1 86.401 0.0037 9.60
4-1 80 70 10 0.0250 1 86.401 0.0036 9.29 9.29
4-2 80 70 10 0.0250 1 86.401 0.0035 9.08
4-3 80 70 10 0.0250 1 86.401 0.0037 9.51
5-1 80 70 10 0.0333 1 86.401 0.0036 9.42 9.22
5-2 80 70 10 0.0333 1 86.401 0.0035 9.08
5-3 80 70 10 0.0333 1 86.401 0.0035 9.1 6
6-1 80 70 10 0.041 7 1 86.401 0.0036 9.29 9.36
6-2 80 70 10 0.041 7 1 86.401 0.0036 9.29
6-3 80 70 10 0.041 7 1 86.401 0.0037 9.51
7-1 80 70 10 0.0500 1 86.401 0.0035 9.1 8 9. 1 5
7-2 80 70 io 0.0500 1 86.401 0.0035 9.08
7-3 80 70 10 0.0500 1 86.401 0.0035 9. 1 8
8-1 80 70 10 0.061 7 1 86.401 0.0035 9.08 9.1 9
8-2 80 70 10 0.06 1 7 1 86.401 0.0036 9.42
8-3 80 70 10 0.06 1 7 1 86.401 0.0035 9.08
1 -1 •1 00 70 30 0.01 00 41 .888 0.0077 6.63 6.70
1 -2 1 00 70 30 0.01 00 41 .888 0.0078 6.70
1 -3 1 00 70 30 0.01 00 41 .888 0.0078 6.77
2-1 1 00 70 30 0.0133 41 .888 0.0080 6.92 6.95
Appendix I 218
2-2 1 00 70 30 0.01 33 41 .888 0.0081 6.99

2-3 1 00 70 30 0.0133 41 .888 0.0080 6.95
3-1 100 70 30 0.0250 41 .888 0.0083 7. 1 3 7.1 1
3-2 1 00 70 30 0.0250 41 .888 0.0083 7.20
3-3 1 00 70 30 0.0250 41 .888 0.0081 6.99
4-1 1 00 70 30 0.0333 41 .888 0.0085 7.35 7.35
4-2 1 00 70 30 0.0333 41 .888 0.0085 7.35
4-3 1 00 70 30 0.0333 41 .888 0.0085 7.35
5-1 1 00 70 30 0.041 7 41 .888 0.0086 7.42 7.35
5-2 1 00 70 30 0.04 1 7 41 .888 0.0085 7.35
5-3 1 00 70 30 0.041 7 41 .888 0.0084 7.28
6-1 1 00 70 30 0.0500 41 .888 0.0084 7.28 7.32
6-2 1 00 70 30 0.0500 41 .888 0.0085 7.35
6-3 1 00 70 30 0.0500 41 .888 0.0085 7.35
7-1 1 00 70 30 0.061 7 41 .888 0.0085 7.35 7.37
7-2 1 00 70 30 0.06 1 7 41 .888 0.0086 7.42
7-3 1 00 70 30 0.06 1 7 41 .888 0.0085 7.35
Hcal-FF-RS-LlT
TRIALS ST EV LlT FF RS W8XD Heal Ave-Heal
(OC ) (OC ) (K) . (xl O·3 m3/s) (rad/s) (x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 60 40 20 0.0500 1 86.401 0.0057 7.63 7.74
1 -2 60 40 20 0.0500 1 86.401 0.0059 7.91
1 -3 60 40 20 0.0500 1 86.401 0.0058 7.69
2-1 70 50 20 0.0500 1 86.401 0.0062 8.22 8.1 5
2-2 70 50 20 0.0500 1 86.401 0.0061 8.05
2-3 70 50 20 0.0500 1 86.401 0.0062 8.1 6
3-1 80 60 20 0.0500 1 86.401 0.0067 8.73 8.84
3-2 80 60 20 0.0500 1 86.401 0.0067 8.79
3-3 80 60 20· 0.0500 1 86.401 0.0069 9.01
4-1 90 70 20 0.0500 1 86.401 0.0076 9.88 9.83
4-2 90 70 20 0.0500 1 86.401 0.0076 9.93
4-3 90 70 20 0.0500 1 86.401 0.0074 9.67
. . .
2 3 The expenmental resu ts 0btamed m
' C entntherm WIt sugar so 1utlOns
For 20% sugar solution
Heal-LlT
TRIALS ST EV 6T FF RS W exp Heal Ave-Heal
(OC ) (OC ) (K) (x1 0·3m 3/s) (rad/s) (x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 65 60 5 0.0500 1 86.401 0.001 3 6.99 7.1 3
1 -2 65 60 5 0.0500 1 86.401 0.001 4 7.42
1 -3 65 60 5 0.0500 1 86.401 0.0013 6.99
2-1 70 60 10 0.0500 1 86.401 0.0028 7.42 7.71
2-2 70 60 10 0.0500 1 86.401 0.0030 7.75
2-3 70 .
60 10 0.0500 1 86.401 0.0030 7.97
3-1 75 60 15 0.0500 1 86.401 0.0047 8.22 8.03
3-2 75 60 15 0.0500 1 86.401 0.0046 8.01
3-3 75 60 15 0.0500 1 86.401 0.0045 7.86
4-1 80 60 20 0.0500 1 86.401 0.0063 8.24 8.24
Appendix I 219
4-2 80 60 20 0.0500 1 86.401 0.0063 8. 1 9

4-3 80 60 20 0.0500 1 86.401 0.0063 8.30
5-1 85 60 25 0.0500 1 86.401 0.0078 8.21 8.38
5-2 85 60 25 0.0500 1 86.401 0.0082 8.56
5-3 85 60 25 0.0500 1 86.401 0.0080 8.38
6-1 90 60 30 0.0500 1 86.401 0.0096 8.41 8.42
6-2 90 60 30 0.0500 1 86.401 0.0097 8.48
6-3 90 60 30 0.0500 1 86.401 0.0096 8.37
7-1 95 60 35 0.0500 1 86.401 0.01 03 7.74 7.80
7-2 95 60 35 0.0500 1 86.401 0.01 05 7.86
8-1 1 05 60 45 0.0500 1 86.401 0.01 30 7.59 7.73
8-2 1 05 60 45 0.0500 1 86.401 0.01 35 7.88
9-1 1 10 60 50 0.0500 1 86.401 0.01 44 7.53 7.45
9-2 110 60 50 0.0500 1 86.401 0.0141 7.38
Heal-RS
TR IALS ST EV �T FF RS W axp H eal Ave-Heal
.COC) .eCJ (K) (x1 0·3m 3/s) (radis) ix1 0·3m3/s) (kW/m2 K) (kW/m2 K)
1-1 80 70 10 0.0500 41 .888 0.0025 6.48 6.56
1 -2 80 70 10 0.0500 41 .888 0.0025 6.59
1 -3 80 70 10 0.0500 41 .888 0.0025 6.59
2-1 80 70 10 0.0500 73.304 0.0029 7.56 7.31
2-2 80 70 10 0.0500 73.304 0.0029 7.46
2-3 80 70 10 0.0500 73.304 0.0027 6.92
3-1 80 70 10 0.0500 1 04.720 0.0030 7.78 7.92
3-2 80 70 10 0.0500 1 04.720 0.0030 7.78
3-3 80 70 10 0.0500 1 04.720 0.0032 8.21
4-1 80 70 10 0.0500 1 46.608 0.0033 8.43 8.39
4-2 80 70 10 0.0500 1 46.608 0.0033 8.54
4-3 80 70 10 0.0500 1 46.608 0.0032 8.21
5-1 80 70 16 0.0500 1 78.024 0.0033 8.54 8.82
5-2 80 70 10 0.0500 1 78.024 0.0034 8.86
5-3 80 70 10 0.0500 1 78.024 0.0035 9.08
Heal-FF (Fig1 .3. 1 )
TRIALS ST EV �T FF RS W Heal Ave-Heal
(OC ) (OC ) (K) (x 1 0-3m 3/S) 3
(rad/s) (x1 0- m 3/s) (kW/m K)2 (kW/m 2 K)
1 -1 70 60 10 0.01 33 1 46.608 0.0030 7.73 7.96
1 -2 70 60 10 0.01 33 1 46.608 0.0030 7.97
1 -3 70 60 10 0.01 33 1 46.608 0.0031 8.1 9
2-1 70 60 10 0.0250 1 46.608 0.0031 8.1 9 7.97
2-2 70 60 10 0.0250 1 46.608 0.0030 7.86
2-3 70 60 10 0.0250 1 46.608 0.0030 7.86
3-1 70 60 10 0.0333 1 46.608 0.0032 8.30 8.12
3-2 70 60 10 0.0333 1 46.608 0.0031 8.21
3-3 70 60 10 0.0333 1 46.608 0.0030 7.86
4-1 70 60 10 0.041 7 1 46.608 0.0032 8.30 8.26
4-2 70 60 10 0.0417 1 46.608 0.0031 8.08
4-3 70 60 10 0.041 7 1 46.608 0.0032 8.41
5-1 70 60 10 0.0500 1 46.608 0.0031 8.1 9 8.26
5-2 70 60 10 0.0500 1 46.608 0.0032 8.30
5.3 70 60 10 0.0500 1 46.608 0.0032 8.30
Appendix I 220
6-1 70 60 10 0.061 7 1 46.608 0.0031 8.08 8.37

6-2 70 60 10 0.061 7 1 46.608 0.0032 8.41
6-3 70 60 10 0.061 7 1 46.608 0.0033 8.62
For 30-60% sogar solutions
Hcal-LlT-TS
TRIALS ST EV LlT FF RS Wexo Heal Ave-Heal
(OC ) (OC ) (K) (x1 0·3m3/s) (rad/s) ix1 0·3 m3/s) (kW/m2 K) (kW/m2 K)
30%
1 -1 75 70 5 0.041 7 1 86.401 0.0013 7.86 7.53
1 -2 75 70 5 0.0417 1 86.401 0.001 3 7.37
1 -3 75 70 5 0.041 7 1 86.401 0.001 3 7.37
2-1 80 70 10 0.041 7 1 86.401 0.0030 8.28 8.23
2-2 80 70 10 0.0417 1 86.401 0.0030 8.28
2-3 80 70 10 0.041 7 1 86.401 0.0030 8.1 4
3-1 85 70 15 0.0417 1 86.401 0.0045 8.1 8 8.13
3-2 85 70 15 0.041 7 1 86.401 0.0044 7.95
3-3 85 70 15 0.0417 1 86.401 0.0046 8.25
4-1 90 70 20 0.0417 1 86.401 0.0060 8.08 8.00
4-2 90 70 20 0.0417 1 86.401 0.0060 8.02
4-3 90 70 20 0.041 7 1 86.401 0.0059 7.91
5-1 95 70 25 0.041 7 1 86.401 0.0078 8.24 8.1 9
5-2 95 70 25 0.041 7 1 86.401 0.0077 8.1 5
5-3 95 70 25 0.041 7 1 86.401 0.0077 8.1 9
40%
1 -1 75 70 5 0.0417 1 86.401 0.001 0 6.27 6.23
1 -2 75 70 5 0.041 7 1 86.401 0.001 0 6.48
1 -3 75 70 5 0.041 7 1 86.401 0.0009 5.94
2-1 80 70 10 0.0417 1 86.401 0.0024 6.92 6.87
2-2 80 70 10 0.041 7 1 86.401 0.0024 6.96
2-3 80 70 10 0.041 7 1 86.401 0.0023 6.72
3-1 85 70 15 0.041 7 1 86.401 0.0038 7.04 6.98
3-2 85 70 15 0.041 7 1 86.401 0.0037 6.79
3-3 85 70 15 0.041 7 1 86.401 0.0038 7. 1 0
4-1 90 70 20 0.041 7 1 86.401 0.0053 7.21 7.1 6
4-2 90 70 20 0.041 7 1 86.401 0.0052 7.05
4-3 90 70 20 0.041 7 1 86.401 0.0053 7.21
5-1 95 70 25 0.041 7 1 86.401 0.0069 7.40 7.36
5-2 95 70 25 0.041 7 1 86.401 0.0068 7.29
5-3 95 70 25 0.041 7 1 86.401 0.0068 7.38
50%
1 -1 75 70 5 0.041 7 1 86.401 0.0007 5.24 5.20
1 -2 75 70 5 0.041 7 1 86.401 0.0006 4.98
1 -3 75 70 5 0.041 7 1 86.401 0.0007 5.37
2-1 80 70 10 0.041 7 1 86.401 0.001 9 5.99 5.99
2-2 80 70 10 0.041 7 1 86.401 0.0020 6.25
2-3 80 70 10 0.041 7 1 86.401 0.001 8 5.73
3-1 85 70 15 0.041 7 1 86.401 0.0032 6.26 6.22
3-2 85 70 15 0.041 7 1 86.401 0.0031 6.08
3-3 85 70 15 0.041 7 1 86.401 0.0033 6.34
4-1 90 70 20 0.041 7 1 86.401 0.0045 6.32 6.26
Appendix I 22 1
4-2 90 70 20 0.0417 1 86.401 0.0043 6.02

4-3 90 70 20 0.041 7 1 86.401 0.0045 6.42
5-1 95 70 25 0.0417 1 86.401 0.0059 6.54 6.51
5-2 95 70 25 0.0417 1 86.401 0.0058 6.40
5-3 95 70 25 0.041 7 1 86.401 0.0059 6.59
60%
1-1 75 70 5 0.041 7 1 86.401 0.0002 1 .85 1 .70
1 -2 75 70 5 0.041 7 1 86.401 0.0001 1 .54
1 -3 75 70 5 0.041 7 1 86.401 0.0001 1 .70
2-1 80 70 10 0.041 7 1 86.401 0.001 1 3.96 3.96
2-2 80 70 10 0.041 7 1 86.401 0.001 2 4.26
2-3 80 70 10 0.041 7 1 86.401 0.001 0 3.65
3-1 85 70 15 0.041 7 1 86.401 0.0023 5.00 4.91
3-2 85 70 15 0.041 7 1 86.401 0.0023 4.91
3-3 85 70 15 0.041 7 1 86.401 0.0023 4.82
4-1 90 70 20 0.0417 1 86.401 0.0033 4.93 4.89
4-2 90 70 20 0.041 7 1 86.401 0.0033 4.93
4-3 90 70 20 0.041 7 1 86.401 0.0032 4.80
5-1 95 70 25 0.041 7 1 86.401 0.0042 4.89 4.97
5-2 95 70 25 0.041 7 1 86.401 0.0043 5.08
5-3 95 70 25 0.041 7 1 86.401 0.0042 4.94
Heal-RS-TS
TRIALS ST EV .1T FF RS W exp Heal Av e-Heal
(O (O
C) C) (K) (x1 0·3m 3/s) (rad/s) (x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
30%
1 -1 80 70 10 0.0417 52.360 0.0024 6.67 6.64
1 -2 80 70 10 0.0417 52.360 0.0023 6.44
1 -3 80 70 10 0.0417 52.360 0.0025 6.81
2-1 80 70 10 0.0417 83.776 0.0026 7.1 3 7.09
2-2 80 70 10 0.041 7 83.776 0.0027 7.36
2-3 80 70 10 0.041 7 83.776 0.0025 6.78
3-1 80 70 10 0.0417 1 04.720 0.0028 7.82 7.78
3-2 80 70 10 0.0417 1 04.720 0.0027 7.36
3-3 80 70 10 0.0417 1 04.720 0.0030 8. 1 6
4-1 · 80 70 10 0.0417 1 57.080 0.0029 7.93 8.01
4-2 80 70 10 0.041 7 1 57.080 0.0030 8.28
4-3 80 70 10 0.041 7 1 57.080 0.0028 7.82
5-1 80 70 10 0.041 7 1 86.401 0.0030 8.28 8.23
5-2 80 70 10 0.0417 1 86.401 0.0030 8.28
5-3 80 70 10 0.0417 1 86.401 0.0030 8.14
40%
1 -1 80 70 10 0.0417 52.360 0.00 1 8 5.28 5.25
1 -2 80 70 10 0.041 7 52.360 0.001 9 5.52
1 -3 80 70 10 0.041 7 52.360 0.001 7 4.95
2-1 80 70 10 0.0417 83.776 0.0020 5.67 5.61
2-2 80 70 10 0.041 7 83.776 0.001 9 5.40
2-3 80 70 10 0.0417 83.776 0.0020 5.76
3-1 80 70 10 0.0417 1 04.720 0.0021 6.00 6.00
3-2 80 70 10 0.041 7 1 04.720 0.0022 6.24
3-3 80 70 10 0.0417 1 04.720 0.0020 5.76
Appendix I 222
4-1 80 70 10 0.041 7 1 57.080 0.0023 6.72 6.68

4-2 80 70 10 0.041 7 1 57.080 0.0024 6.82
4-3 80 70 10 0.0417 1 57.080 0.0023 6.48
5-1 80 70 10 0.041 7 1 86.401 0.0024 6.92 6.87
5-2 80 70 10 0.041 7 1 86.401 0.0024 6.96
5-3 80 70 10 0.04 17 1 86.401 0.0023 6.72
50%
1 -1 80 70 10 0.04 17 52.360 0.001 3 4. 1 7 4. 1 7
1 -2 80 70 10 0.041 7 52.360 0.001 3 3.96
1 -3 80 70 10 0.041 7 52.360 0.00 1 4 4.37
2-1 80 70 10 0.041 7 83.776 0.00 1 4 4.48 4.44
2-2 80 70 10 0.041 7 83.776 0.001 4 4.37
2-3 80 70 10 0.041 7 83.776 0.001 4 4.48
3-1 80 70 10 0.04 17 1 04.720 0.001 6 4.84 4.81
3-2 80 70 10 0.041 7 1 04.720 0.001 5 4.63
3-3 80 70 10 0.041 7 1 04.720 0.001 6 4.95
4-1 80 70 10 0.041 7 1 57.080 0.001 8 5.73 5.55
4-2 80 70 10 0.041 7 1 57.080 0.001 8 5.47
4-3 80 70 10 0.041 7 1 57.080 0.001 8 5.47
5-1 80 70 10 0.041 7 1 86.401 0.001 9 5.99 5.99
5-2 80 70 10 0.041 7 1 86.401 0.0020 6.25
5-3 80 70 10 0.041 7 1 86.401 0.001 8 5.73
60%
1 -1 80 70 10 0.041 7 52.360 0.0007 2.44 2.42
1 -2 80 70 10 0.041 7 52.360 0.0007 2.56
1 -3 80 70 10 0.041 7 52.360 0.0006 2.28
2-1 80 70 10 0.041 7 83.776 0.0008 2.92 2.90
2-2 80 70 10 0.04 1 7 83.776 0.0008 3.04
2-3 80 70 1 0 .. 0.041 7 83.776 0.0008 2.74
3-1 80 70 10 0.041 7 1 04.720 0.001 0 3.47 3.49
3-2 80 70 10 0.041 7 1 04.720 0.001 0 3.65
3-3 80 70 10 0.041 7 1 04.720 0.0009 3.35
4-1 80 70 10 0.041 7 1 57.080 0.001 0 3.53 3.61
4-2 80 70 10 0.04 17 1 57.080 0.001 0 3.65
4-3 80 70 10 0.04 17 1 57.080 0.001 0 3.65
5-1 80 70 10 0.04 17 1 86.401 0.001 1 3.96 3.96
5-2 80 70 10 0.041 7 1 86.401 0.001 2 4.26
5-3 80 70 10 0.041 7 1 86.401 0.001 0 3.65
Sugar 20 and 68%
Hcal-TS
TRIALS ST EV �T FF RS W exP Heal Ave-Heal
(OC ) .(OC ) (K) (x1 0·3m 3/s) 3 3
(rad/s) (x1 0- m /s) (kW/m K)2 (kW/m 2 K)
1 -1 75 70 10 0.041 7 1 86.401 0.0035 9. 1 5 9.25
1 -2 75 70 10 0.041 7 1 86.401 0.0035 9.22
1 -3 75 70 10 0.041 7 1 86.401 0.0035 9.37
6-1 80 70 10 0.041 7 1 86.401 0.0003 1 .22 1 .09
6-2 80 70 10 0.041 7 1 86.401 0.0002 0.97
6-3 80 70 10 0.041 7 1 86.401 0.0003 1 .08
Appendix I 223
2 4 The expenmental resu ts 0btamed In

' Centn' therm Wl· th skim nu' lk ( 1 1 % total solids)
.
H�I-ilT
TRIALS ST EV ilT FF RS W exp H eal Ave-Heal
(OC ) (OC ) (K) ! (x1 0·3m 3/s) (rad/s) (x1 0·3 m%) (kW/m2 K) (kW/m2 K)
1 -1 65 60 5 0.0500 1 86.401 0.0013 6.55 6.88
1 -2 65 60 5 0.0500 1 86.401 0.001 3 6.77
1 -3 65 60 5 0.0500 1 86.401 0.0014 7.34
2-1 70 60 10 0.0500 1 86.401 0.0030 7.86 7,60
2-2 70 60 10 0.0500 1 86.401 0.0029 7.60
2-3 70 60 10 0.0500 1 86.401 0.0028 7.34
3-1 75 60 15 0.0500 1 86.401 0.0045 7.86 7.87
3-2 75 60 15 0.0500 1 86.401 0.0046 8.01
3-3 75 60 15 0.0500 1 86.401 0.0044 7.74
4-1 80 60 20 0.0500 1 86.401 0.0063 8. 1 9 8.1 1
4-2 80 60 20 0.0500 1 86.401 0.0062 8.08
4-3 80 60 20 0.0500 1 86.401 0.0062 8.08
5-1 85 60 25 0.0500 1 86.401 0.0078 8.21 8.27
5-2 85 60 25 0.0500 1 86.401 0.0078 8.21
5-3 85 60 25 0.0500 186.401 0.0080 8.38
6-1 90 60 30 0.0500 1 86.401 0.0096 8.41 8.31
6-2 90 60 30 0.0500 1 86.401 0.0096 8.37
6-3 90 60 30 0.0500 1 86.401 0.0093 8.1 5
Heal-RS
TRIALS ST EV ilT FF RS W exp Heal Ave-Heal
eC ) _eC ) (K) (x1 0·3m 3/s) (rad/s) (x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 80 70 10 0.0500 41 .888 0.0025 6.48 6.56
1 -2 80 70 10 0.0500 41 .888 0.0025 6.59
1 -3 80 70 10 0.0500 41 .888 0.0025 6.59
2-1 80 70 10 0.0500 73.304 0.0029 7.56 7.31
2-2 80 70 10 0.0500 73.304 0.0029 7.46
2-3 80 70 10 0.0500 73.304 0.0027 6.92
3-1 80 70 10 0.0500 1 04.720 0.0030 7.78 7.92
3-2 80 70 10 0.0500 1 04.720 0.0030 7.78
3-3 80 70 10 0.0500 1 04.720 0.0032 8.21
4-1 80 70 10 0.0500 1 46.608 0.0033 8.43 8.39
4-2 80 70 10 0.0500 1 46.608 0.0033 8.54
4-3 80 70 10 0.0500 1 46.608 0.0032 8.21
5-1 80 70 10 0.0500 1 78.024 0.0033 8.54 8.82
5-2 80 70 10 0.0500 1 78.024 0.0034 8.86
5-3 80 70 10 0.0500 1 78.024 0.0035 9.08
Heal-FF
TRIALS ST EV L\T FF RS Wexp Heal Ave-Heal
(OC ) (OC ) (K) I (x1 0·3 m3/s) (rad/s) (x1 0·3m 3/s) (kW/m2 K) (kW/m2 K)
1 -1 70 60 10 0.0 1 33 1 46.608 0.0030 7.75 7.50
1 -2 70 60 10 0.01 33 1 46.608 0.0028 7.21
1 -3 70 60 10 0.01 33 1 46.608 0.0029 7.53
2-1 70 60 10 0.0250 1 46.608 0.0029 7.53 7.71
2-2 70 60 10 0.0250 1 46.608 0.0030 7.86
2-3 70 60 10 0.0250 1 46.608 0.0030 7.75
3-1 70 60 10 0.0333 1 46.608 0.0030 7.86 7.86
Appendix I 224
3-2 70 60 10 0.0333 1 46.608 0.0029 7.64

3-3 70 60 10 0.0333 1 46.608 0.0031 8.08
4-1 70 60 10 0.041 7 1 46.608 0.0031 8.1 2 7.98
4-2 70 60 10 0.041 7 1 46.608 0.0031 8.08
4-3 70 60 10 0.041 7 1 46.608 0.0030 7.75
5-1 70 60 10 0.0500 1 46.608 0.0031 8.03 8.03
5-2 70 60 10 0.0500 1 46.608 0.0031 8.1 9
5-3 70 60 10 0.0500 1 46.608 0.0030 7.8 6
6-1 70 60 10 0.061 7 1 46.608 0.0031 8.1 9 8.1 9
6-2 70 60 10 0.0617 1 46.608 0.0031 8.08
6-3 70 60 10 0.061 7 1 46.608 0.0032 8.30
Appendix n
Calculation of surfaces areas on the cone
The overall heat transfer rate (Q) from the heating medium to evaporating liquid across
the wall of the cone and the liquid films can be described by the following equation:
So, the temperature differences can be calculated as follows:
The overall heat transfer equation based on inner surface of the cone can be written as
follows :
The surface areas on the cone:

Heat Transfer and Fouling in Film Evaporator with Rotating Surfaces 226
Appendix ll
Correlations of the physical properties for water, sugar solution

and skim milk
3. 1 Water physical properties
The data of physical properties for water were taken from Thermophysical Properties of
Water Substance by Rogers and Mayhew ( 1982). A computer programm of Fig.P (Fig.P
Software Corporation, Ver.6.0, 1992) was used to fit the data and generate the correlations.
These correlations could be used for the temperature range of 40 to 1 10°C.
3.1.1 Water density
Water density as a function of temperature was determined by the following correlation in the
units ofkglm3 :
p = 1 . 0167x 103 - 0.9143T + 1 . 0929x 102

-8. 675x 10·5 r3 + 1. 01x10-1 1r6
(R2 = 99.95%, df = l O)
3.1.2 Water kinematic viscosity
Water dynamic viscosity as a function of temperature was determined by the following

correlation in the units of m3/s:
73.87 )
v = 1.407 x 10 exp (
-7 - T-
(R2 = 99. 1%, df = 1 6)
3.1.3 Water thennal capacity
Water thermal capacity as a function of temperature was determined by the following

correlation in the units of J/kg.K:
Cp = 4.1828 X 1 0 3 - 0.4457 T + 7.959 X 1 0 -3 T2

(R2 = 99.9%, df = 1 6)
Appedix III 227
3.1.4 Water thennal conductivity
Water thennal conductivity as a function of temperature was detennined by the following

correlation in the units of W/mK:
6
k = 0.5682 + 1.85 x 10-3 T - 7.273 X 10- T 2
(R2 = 99.9%, df = 1 6)
3.1.5 Latent heat of vaporisation
Latent heat of vapourization as a function of evaporation temperature was detennined by the

following correlation in the units of J/kg:
hlg =
2.5 1 32 X
6
1 0 - 2.565 X 1 0 3 Tevp
(R2 = 99.9%, df= 1 6)
3.2 Sugar solution physical properties
3.2.1 Sugar solution density
Sugar solution density as a function of temperature and total solids was detennined by the
3
following correlation in the units of kglm (MacKereth, 1993):
p ss =
(0.983 1 - 9.667 x 10-3 T · C1 + 6.1 87 X 10 -4 C1 . C2 5
- 6.2268 X 10- Cl 2
+ 4.095 X 1 0 -4 c ) 2 I?20
Where:
D20 = 1 1 83.2 + 2 1 6.147 C3 + 30.9247 C/ + 0.27466 C/
C1 =
T
__-__60
'
C2 =
TS - 50
' C3 =
TS - 41.5
20 20 4 1.5
The total solids is in the range of 30 to 70% and the temperature is in the range of 40 to
80°C.
3.2.2 Sugar solution dynamic viscosity
The correlation was proposed by Campanella (1992) and the application ranges are of 0
t060% total solids and 0-95°C temperature. The unit of viscosity is in centipoise.
Il ss =
(
8. 194 x 10-2 exp 9.176 x 10-2 (1 + TS)4. 9029 )
exp
( 2.5 027 x 1 0
7 + 4.44 33 (1 + TS) 4. 9
029 )
T 2 .8774
------ - ------- ---�---
Appedix III 228
3.2.3 Sugar solution thennal conductivity
Sugar solution thenna! conductivity as a function of temperature and total solids was
determined by the following correlation, which fitted to 9 data points from Honig ( 1963)
using Fig.P programm. The units of thennal conductivity are W/mK:
kss = 0.596 + 9.3 1 x 1 0 -4 T - 3. 49 X 10-3 TS

(R2 = 99.7%)
The temperature is in the range of 50 to 70°C and total solids are in the range of 40 to 60% .
3.2.4 Sugar solution boiling point elevation
The boiling point of skim milk as a function of total solids was determined by the following
correlations in the unit of degree centigrade (Mechereth, 1993):
BPEss = ( 2 -4
exp -9.013 x 10-2 - 3.33 X 10- TS + 8.095 X 1 0 TS 2 )
3.3 Skim milk physical properties
3.3.1 Skim milk density
Skim milk density as a function of temperature and total solids was determined by the
3
following correlation in the units of kg/m (Hall and Hedrick, 1 966):
�
(2K6 Ts - K4 T - K5 ) 2 - (K4 T - K5 ) 2 + 4K6 (K1 - K2 T + K3 T 2 )

P skim =
4
K6
Where: Kl = 989; K = 0.064; K3 = 0.0024; � = 0.0076; Ks = 3.75; Kt; = 0.001 66

2
The temperatures are in the range of 0 to 10°C and total solids are in the range of 9 to 50%.
3.3.2 Skim milk dynamic viscosity
The correlation fitted to 6 data point from Wood (1982) using Fig.P programm. The
dynamic viscosity was in the unit of centipoise. The temperature is in the range of 30-70°C.
11 skim = 14.689g exp ( -O.0949T) + 0.7286
(R2 = 98.6%)
Appedix III 229
3.3.3 Skim milk thennal conductivity
The correlation fitted to 6 data point from Wood ( 1982) using Fig.P prograrnrn.
The thermal conductivity was in the units ofW/mK The temperature is in the range of 40 to
90°C.
.'
t
ks",m =
0.54 + 2.0 x 1 0 -3 T - 1.07 X 1 0 -5 T 2
(R2 = 99.7%)
3.3.4 Skim milk boiling point elevation
The boiling point of skim milk as a function of total solids was determined by the following
correlations in the unit of degree centigrade (Mackereth, 1993):
BPEskirn =
2.78 X 1 0 -3 TS + 2.94 X 1 0 -4 TS 2 - 4.65 X 1 0 -3
(R2 = 98.5% )
Appendix IV
Results of numerical calculation
4. 1 The theoretical results of the model
"
x 1 00 500 1 000 5000 1 0000
TI (0C) 70 70 70 70 70
FF (x1 O·s m%) 5.0 5.0 5.0 5.0 5.0
RS(radls) 1 57 1 57 1 57 1 57 1 57
�=1 00 48.24 23.84 1 6.54 6.28 4.02
�=200 33.34 1 3.84 9.03 3.1 8 2.04

"
0 �=300 24.78 9.43 6.03 2.09 1 .32
�=400 1 9. 1 3 6.96 4.42 1 .52 0.96
�=500 1 5.04 5.33 3.38 1 .1 6 0.73
X 1 00 500 1 000 5000 1 0000

TI (0C) 70 70 70 70 70
� 40° 40° 40° 40° 40°

RS(radls) 1 57 1 57 1 57 1 57 1 57
'"
FF=6.7x1 0·s (m3/s) 21 .09 7.67 4.87 1 .68 1 .06

0" FF=6.7x1 0·s (m3/s) 1 9. 1 3 6.96 4.42 1 .52 0.96
FF=6.7x1 0.5 (m3/s) 1 6.66 6.06 3.85 1 .33 0.84
FF=6.7x1 0·s (m%) 1 3.35 4.85 3.08 1 .06 0.67
x" (X1 03) 5.57 5.57 5.57 5.57 5.57 5.57

TI �C) 70 70 70 70 70 70
FF (x1 0·s m3/s) 5.0 5.0 5.0 5.0 5.0 5.0
Ro 4x1 0·s 4x1 0-4 4x1 0·3 4x1 0·2 4x1 0·1 4
�= 100 1 2.54 9.07 4.70 2.20 1 .00 0.48
�=20o 1 0.00 7.20 3.70 1 .75 0.80 0.37

" 0.71 0.33
0 �=30o 8.69 6.30 3.25 1 .53
�=40o 7.88 5.70 2.95 1 .39 0.65 0.30
�=50o 7.29 5.27 2.73 1 .28 0.60 0.28

Appendix IV 23 1
X· (X1 03) 5.57 5.57 5.57 5.57 5.57 5.57

TI �C) 70 70 70 70 70 70
FF (x1 0·s m%) 5.0 5.0 5.0 5.0 5.0 5.0
RS(rad/s) 52 1 05 1 57 21 0 262 314
�= 1 0o 0.097 0.1 35 0.1 71 0.204 0.235 0.265
�=20o 0. 1 71 0.260 0.337 0.407 0.472 0.532

Nu � =30o 0.253 0.394 0.51 4 0.622 0.721 0.81 4
�=40o 0.344 0.540 0.706 0.855 0.991 1 .1 1 9
�=50o 0.449 0.708 0.927 1 .1 22 1 .302 1 .470
X (m) 0.093 0.1 24 0.1 55 0.1 86 0.21 7

ET �C) 70 70 70 70 70
6T (K) 10 10 10 10 10
FF (x1 0·Sm3/s) 5.0 5.0 5.0 5.0 5.0
RS(rad/s) 1 57 1 57 1 57 1 57 1 57
kJdw (kw/m2 . K) 28.1 3 28.1 3 28.1 3 28.1 3 28.1 3
hl¢(kW/m2 .K) 1 3.72 1 6.61 1 9.27 21 .76 24.1 2
hs (kW/m2 .K) 60.91 59.14 57.88 56.66 55.80
Heal (kW/m2 .K) 7.96 8.83 9.50 10.03 1 0.48
X (m) 0.093 0.1 24 0.1 55 0.186 0.2 1 7

ET (0C) 70 70 70 70 70
6T (K) 10 10 10 10 10
FF (x1 0·sm%) 5.0 5.0 5.0 5.0 5.0
RS(rad/s) 1 86 1 86 1 86 186 1 86
Ty,1 �C) 75.83 75.34 74.96 74.64 74.37
Tw.; �C) 78.68 78.51 78.38 78.23 78.1 2
o (x1 0·�(m) 4.38 3.70 3.25 2.91 2.65

Re 345 257 205 170 1 46
Appendix IV
23 2
4.2. The calculated heat transfer coefficients in the Centrithenn evaporator
4.2. 1 For water
ET (0C) 60 70 70 70 70
.:lT (K) 50 10 10 10 10
FF (m3/s) 5.0x10'5 0.7x10,5 1 .0x1 0,5 1 .3x1 0,5 2.5x1 0,5
RS(radls) 1 86 1 86 1 86 1 86 1 86
TYA (0C) 81 .5 72.7 73.1 73.5 74.2
Tw.; (0C) 98.0 77.4 77.4 77.7 78.0
hi '(X1 04) 2.21 4.90 3.88 3.40 2.50
0Nlm2.K)
hs (X1 04) 3.85 5.1 0 5.1 0 5.30 5.50
0Nlm2 .K)
HcaJ(x 1 03) 9.1 8 1 3.22 1 2.33 1 1 .89 10.68
0Nlm2.K)
ET (0C) 70 70 70 70 70
�T (K) 10 �
10 10 10 30
FF (m3/s) 3.3x1 0,5 4.2x1 0.5 5.0x1 0'5 6.2x1 0.5 1 .0x1 0'5
RS(radls) 1 86 1 86 1 86 1 86 42
TYA tc) 74.5 74.7 74.9 75.0 81 .6
Tw.; tc) 78.1 78.2 78.3 78.2 88.6
hl'(x1 04) 2.28 2.08 1 .95 1 .80 1 .70
0Nlm2 .K)
hs (X1 04) 5.6 5.62 5.72 5.65 1 .78
0Nlm 2.K)
HcaJ(x 1 03) 1 0.22 9.81 9.54 9.15 6.64
(W/m2.K)
Appendix IV
233
ET (0C) 70 70 70 70 70
�T (K) 30 30 30 30 30
FF (m%) 1 .3x1 0·5 2.5x1 0-5 3.3x1 0-5 4.2x1 0-5 5.0x1 0-5
RS(radls) 42 42 42 42 42
T\\1 tC) 83.1 85.7 86.8 87.6 88. 1
Tws tc) 89.8 91 .4 97.2 92.8 93.1
hl'(x104) 1 .40 1 .00 0.90 0.80 0.77
0Nlm2.K)
hs (X104) 1 .80 1 .90 1 .96 1 .99 2.0
(W/m2.K)
Hcal(x1 03) 6.23 5.36 5.06 4.81 4.63
0Nlm2.K)
ET tc) 70 60 60 60 60
�T (K) 10 10 10 10 10
.
FF (m%) 6.2x1O-5 r 1 .3x1 0-5 2.5x1O-5 3.3x 1 0-5 4.2x1 0 5
RS(radls) 42 1 47 147 1 47 1 47
T\\1 (0C) 88.7 64.0 64.5 65.0 65.1
Tws (0C) 93.4 67.8 68.0 68.2 68.1
hl'(x1 04) 0.71 2.67 2.00 1 .81 - 1 .65
0Nlm2.K)
hs (X104) 2.00 4.57 4.53 4.82 4.75
0Nlm2.K)
Hcal(x103) 4.43 1 0.49 9.25 8.91 8.49
0Nlm2.K)
Appendix IV 234
ET �C) 60 60
�T (K) 10 10
F F (m3/s) 5.0x1 0·5 6.2x1 0.5
RS(rad/s) 1 47 147
T\'.1 �C) 65.2 65.5
Tws (0C) 68.1 68.3
hJ '(x1 04) 1 .55 1 .44
0Nlm2.k)
hs (X104) 4.73 4.89
0Nlm2.K)
HcaJ(x103) 8.20 7.92
0Nlm2.K)
4.2.2 For sugar solutions
TS (%) 20 20 20 20 20
ET �C) 60 60 60 60 60
�T (K) 10 10 10 10 10
FF (m%) 1 .3x1 0·5 �
2.5x1 0·5 3.3x1 0·5 4.2x1 0.5 5.0x1 0·5
RS(rad/s) 1 47 147 147 147 1 47
T\'.1 �C) 64.5 65. 1 65.4 65.6 65.7
Tws �C) 68.0 68.1 68.3 68.4 68.3
hJ'(x1 04) 2.20 1 .60 1 .50 1 .40 1 .30
0Nlm2.K)
hs (X1 04) 4.75 4.79 4.89 4.93 4.90
0Nlm2.K)
HcaJ(x103) 9.81 8.60 8.17 7.85 7.50
0Nlm2.K)
Appendix IV
235
TS (%) 20 20 20 20 20
ET (,C) 60 70 70 70 70
ilT (K) 10 5 10 15 20
FF (m3/s) 6.2x1 0·s 4.2x1 0·s 4.2x1 0-s 4.2x10·s 4.2x1 0.5
RS(rad/s) 1 47 1 86 1 86 186 1 86
TYA (,C) 65.9 73.0 75.5 77.9 80.2
Tw.; ('C) 68.4 74.4 78.5 82.5 86.4
h,'(x104) 1 .20 1 .70 1 .75 1 .79 1 .83
(W/m2.K)
hs (X1 04) 4.99 7.38 5.92 5.29 4.90
(W/m2.K)
Hcat(x103) 7.1 9 9.24 9.07 9.02 9. 1 0
(W/m2.K)
TS (%) 20 20 30 30 30
ET (DC) 70 70 70 70 70
ilT (K) 25 30 5 10 15
4.2x1 0.5 4.2x1 0.5
�
FF (m%) 4.2x1 0-5 4.2x10-5 4.2x1 0-S

RS(rad/s) 1 86 186 1 86 1 86 1 86
TYA (DC) 82.4 84.5 73.0 75.8 78.3
Tw.; ('C) 90.0 94.0 74.2 78.6 82.6
h,'(x 104) 1 .88 1 .92 1 .51 1 .55 1 .58
(W/m2.K)
hs (X1 04) 4.60 4.40 6.80 6.00 5.30
(W/m2.K)
Hcat(x1 03) 9.00 9.01 8.54 8.51 8.46
(W/m2.K)
Appendix I V 236
TS (%) 30 30 30 40 40
ET (0C) 70 70 70 70 70
L\T (K) 20 25 30 5 10
.
FF (m%) 4.2x1 0-5 4.2x1 0 5 4.2x1 0-5 4.2x1 0-5 4.2x1 0.5
RS(rad/s) 1 86 1 86 1 86 186 1 86
T\\4 tC) 80.8 83.0 85.4 73.3 76.2
Tw.; tC) 86.6 90.3 94.3 74.4 78.6
hJ'9x 1 04) 1 .62 1 .66 1 .70 1 .28 1 .31
ry./lm2 .K)
hs (X1 04) 4.97 4.62 4.46 7.29 6.1
(W/m2.K)
HcaI(x 1 03) 8.48 8.46 8.52 7.80 7.75
ry./lm2.K)
TS (%) 40 40 40 40 50
ET tC) 70 70 70 70 70
L\T (K) 15 20 25 30 5
.
FF (m%) 4.2x1 0-5 4.2x1 0-5 4.2x1 0 5 4.2x1 0-5 4.2x1 0-5
RS(rad/s) 1 86 1 86 1 86 1 86 1 86
T\\4 (0C) 79.0 81 .7 84. 3 86.7 73.7
Tw.; (0C) 82.9 87.0 91 .0 94.8 74.5
hJ'(x1 04) 1 .34 1 .38 1 .41 1 .44 1 .03
ry./lm2. K)
hs (X1 04) 5.50 5.1 2 4.83 4.57 7.95
(W/m2.K)
HcaI(x1 03) 7.76 7.79 7.8 2 7.85 6.80
(W/m2 .K)
Appendix IV 237
TS (%) 50 50 50 50 50
ET (,C) 70 70 70 70 70
flT (K) 10 15 20 25 30
FF (m%) 4.2x10·s 4.2x1 0·s 4.2x1 0·s 4.2x1 0·s 4.2x1 0·s
RS(rad/s) 1 86 1 86 1 86 1 86 1 86
T\'A ('C) 77.0 80.0 82.8 85.7 88.5
Tv.s ('C) 79.1 83.3 87.3 91 .5 95.7
hi '(x1 04) 1 .06 1 .09 1 .1 1 1 .1 4 1 .1 7
rN/m2. K)
hs (X1 04) 6.79 5.86 5.28 5.00 4.77
rN/m2.K)
HcaI(x1 03) 6.86 6.86 6.88 6.94 7.00
rN/m2. K)
TS (%) 60 60 60 60 60
ET ('C) 70 70 70 70 70
flT (K) 5 10 15 20 25
4.2x1 0.5 4.2x1 0.5 4.2x1 0.5
,
FF (m3/s) 4.2x1 0.5 4.2x1 0.5

RS(rad/s) 186 1 86 1 86 1 86 1 86
T\'A (0C) 74.2 77.7 81 . 1 84.3 87.5
Tv.s (0C) 74.7 79.2 83.6 87.9 92.2
hl'(x1 04) 7.79 8.00 8.20 8.50 8.70
(W/m2.K)
hs (X1 04 ) 8.92 7.00 6.20 5.60 5.30
rN/m2.K)
Hcal(x1 03) 5.67 5.69 5.74 5.80 5.86
rN/m2.K)
Appendix IV 238
TS (%) 60 68 20 20 20
ET �C) 70 70 70 70 70
�T (K) 30 10 5 10 15
FF (m%) 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5
RS(radls) 1 86 1 86 147 147 1 47
TvA (0C) 90.7 78.6 73.1 75.5 78.2
Tw.; �C) 96.6 79.6 74.4 78.0 82.3
h,'(x104) 8.90 6.1 4 1 .46 1 .48 1 .52
0Nlm2 .K)
hs (X104) 5.1 0 8.23 6.38 4.90 4.5
(W/m2.K)
Hcat(x103) 5.94 4.72 8.23 8.07 8.08
0Nlm2.K)
TS (%) 20 20 20 30 30
ET �C) 70 70 70 70 70
�T (K) 20 25 30 5 10
F F (m%) 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5
RS(radls) 1 47 1 47 147 1 47 1 47
TvA �C) 80.7 83. 1 85.3 73.2 76.0
Tw.; (0C) 86.3 90. 1 93.7 74.3 78.4
h,'(X104) 1 .55 1 .59 1 .62 1 .28 1 .30
0Nlm2.K)
hs (X104) 4.30 4.07 3.87 6.33 5.22
0Nlm2.K)
Hcat(X103) 8.08 8.09 8.09 7.70 7.60
0Nlm2.K)
Appendix IV 239
TS (%) 30 40 40 50 50
ET fC) 70 70 70 70 70
ilT (K) 30 10 30 10 30
FF (m%) 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5 4.2x1 0.5
RS(radls) 1 47 147 1 47 147 147
T'hI (0C) 86.3 76.6 87.6 n.1 89.5
T (0C)
w.; 94.3 78.9 94.9 79.0 95.9
hl'(x104) 1 .44 1 .1 2 1 .14 0.90 0.99
0Nlm2.K)
hs (X1 04) 3.96 5.62 1 .22 5.70 4.30
(W/m2.K)
HcaJ(x103) 7.64 6.96 7.02 6.05 6.24
(W/m2.K)
4.2.3 For skim milk
ET fC) 60 60 60 60 60 70
ilT (K) 10 10 10 10 10 10
FF (xlO·5 m3/s) 1 .3 ,
2.5 3.3 4.2 5.0 4.2
RS(radls) 1 47 147 1 47 1 47 1 47 1 47
T'hI fC) 64.5 65.3 65.5 65.7 65.9 66
T fC)
w.; 67.8 68.3 68.3 68.4 68.5 68.4
hl'(x104) 2. 1 0 1 .60 1 .44 1 .30 1 .24 1.15
(W/m2.K)
hs (X104) 4.60 4.90 4.89 4.93 5.00 4.99
0Nlm2.K)
HcaJ(x1 03) 9.52 8.40 7.91 7.55 7.29 6.96
0Nlm2.K)
Appendix IV 240
4.3 The calculated heat transfer coefficients in the cone evaporator
ET (0C) 70 70 70 70 70
�T (K) 10 10 10 10 10
F F (m%) 5xl O-s 5x1 0-s 5x1 0-5 5xl O-5 5x1 0-s
RS(rad/s) 52 84 1 05 1 57 1 86
Twi (0C) 76.5 76.2 76.0 75.6 75.5
Tws (0C) 77.7 77.7 77.6 77.6 77.7
hl'(x104) 0.51 0.68 0.78 1 .02 1 .12
0Nlm2.K)
hs (x1 0") 1 .46 1 .80 2.1 0 2.50 2.67
0Nlm2.K)
HcaI(x1 03) 3.33 4.20 4.70 5.75 6. 1 3
0Nlm2.K)
ET tc) 70 70 60 60 60
�T (K) 10 10 10 10 10
F F (m%) 5xl O-5 "
5xl O-5 5x1 0-5 5x1 0-5 5xl O-5
RS(rad/s) 209 230 52 84 1 05
Twi (0C) 75.3 75.2 66.6 66.3 76.0
Tws tC) 77.6 77.6 67.7 67.7 67.6
hl'(x 1 04) 1 .24 1 .33 0.48 0:64 0.73
0Nlm2 .K)
hs (X1 04) 2.91 3.05 1 .40 1 .78 1 .96
0Nlm2 .K)
HcaI(X1 03) 6.61 6.92 3.1 7 4.00 4.46
(W/m2 .K)
Appendix IV 24 1
ET rC) 60 60 60 60 70
�T (K) 10 10 10 10 4
FF (m%) 5x1 0·5 5x1 0·5 5x10·5 5x1 0·5 4.2x1 0.5
RS(radls) 1 57 1 86 209 230 230
ToM (0C) 65.7 65.6 65.4 65.3 72.7
Tw.; (0C) 67.6 67.8 67.6 67.6 73.2
hl'(x1 04) 0.96 1 .04 1.17 1 .24 1 .30
0Nlm2.K)
hs (X1 04) 2.40 2.59 2.79 2.93 3.90
0Nlm2.K)
HcaJ(x103) 5.49 5.85 6.32 6.62 7.40
0Nlm2.K)
ET eC) 70 70 60 60 60
�T (K) 10 15 10 10 10
FF (m3/s) 4.2x1 0.5 4.2x1 0.5 1 .7x1 0·5 4.2x1 0.5 6.7x 1 0·5
RS(radls) 230 230 230 230 230
ToM (0C) 75.1 77.2 64.0 65.0 65.7
Tw.; (0C) 76.7 81 .0 66.9 67.4 68.0

"
hl'(x 1 04 ) 1 .40 1 .50 2.1 0 1 .33 1 .1 1

(W/m2.K)
hs (X1 04) 3.1 0 2.70 2.80 2.85 3.04
(W/m2.K)
HcaJ(x103) 7.20 7. 14 8.26 6.82 6.27
(W/m2.K)
Appendix IV 242
ET �C) 50 50 50
LlT (K) 10 10 10
FF (m3;s) 5x1 0's 5x1 0's 5x1 0's
RS(rad/s) 1 57 1 57 1 57
TYA �C) 55.9 65.7 75.6
Tws �C) 57.7 67.6 77.6
hl\x104) 8.70 0.96 1 .02
0Nlm2.K)
hs (X1 04) 2.30 2.40 2.50
0Nlm2.K)
HcaI(x 1 03) 5.16 5.49 5.75
0Nlm2.K)
Heat Transfer and Fouling in Film Evaporator with Rotating Surfaces 243
Appendix V
An example of determining the uncertainty
for overall heat transfer coefficient
The measured overall heat transfer coefficient can be calculated using following equation:
Wexp hfg
Hexp = --'---=-
A I1T
If the evaporating temperature was 70°C ±O.5°e and the steam temperature was 80°C ±O.5
°e, then the temperature difference and its uncertainty can detennined as follows:
2 2
I1T = ( 80 - 70 ) ± ( 0.5 + 0.5 ) 112 = 10 ± 0.7 1 °e
Therefore, 95% confidence interval in percentage terms as follows:
PI = ( 0.7 1/10 ) x 100 = 7. 1
The condensate flow rate was measured in a typical run as 175, 1 85, 188 rnVrnin, so its mean
and uncertainty were 1 82.67 ± 1 1 mlIrnin [Standard deviation ( s ) = 2.55, t = 4.3, 95%
".
confidence interval ( B ) = s · t = 2.55 x 4.3 = 10.965 ] .
Therefore: P2 ( 10.96511 82.67 ) x 1 00 = 6.0

=
2 2 112 = ( 7. 1 2 + 6.02 ) 112

Pr = ( P1 + P2 )
= ( 50.4 + 36 ) 112 = 9.3
The overall heat transfer coefficient was calculated as follows:
Hcal =
1 82.67 x 2360
0.09 X 10
= 7.06 (kW 1m 2 K)
The uncertainty of overall heat transfer coefficient can be calculated as follows:
Heal = 7.06 ± 7.06 x 9.3 1 100 = 7.06 ± 0.66 kW/m2K (or 7.06 kW/ntK ±l O%)
So the 95% confident interval of measured overall heat transfer coefficient was about 1 0%.
Appendix VI
Polyacrylamide gel electrophoresis (pAGE)
Polyacrylamide gel electrophoresis (pAGE) is regarded as a standard method to identify the
undenatured whey proteins in milk. The Bio-Rad Mini Protean II equipment (Bio-Rad
Laboratories, Richmond, CA 94804, USA) was used to perform PAGE. The prepared whey
protein solutions were diluted ( 1 :7.5, 1 :22.5, 1 :45 or 1 :85 for 1 0, 30, 60, or 120glkg WP
solutions respectively) with the appropriate sample buffer and analyzed using native-PAGE as
described by Andrews ( 1983). 10 JlI of each sample was applied to the gel. The gels were
run at 200 V for about 45 minutes until the dye disappeared from the bottom of the gel.
Afterwards the gels were stained for one hour, with Coornassie Blue dye solution in a closed
container with continuous vibrations. This was followed by two destaining steps of one hour
and nineteen hours, in destaining solution. Finished gels were kept in distilled water until
scanning. A 35 rnrn camera fitted with both a green (XI) and an orange (G) Hoya filter (to
minimize the stray light) was used to photograph the gels on a lOOASA Kodak T-max :film.
.'
The gels were scanned using an Ultrascan XL laser densitometer and the results were
analyzed using an LKB 2400 GelScanXL software program (LKB Produkter AB,
Stockholm-Brornma 1, Sweden) to obtain quantitative results. The protein bands on the
resultant gels were identified by concurrent electrophoresis of samples of BSA, u
lactalbumin, �-lactoglobulin A, �-lactoglobulin B. The peak area of each protein band was
reported as a percentage of the corresponding band in the unheated control whey protein
solution samples.
Appendix VII
Experimental results of fouling in the Centritherm evaporators
Time ST ET FT FF Wexp RS H eal Ave-H eal Rf

(hr) (OC ) (OC ) ( OC ) (x1 0-3 (x1 0-3 (rad/s ) (kW/m K) (kW/m2K) (mZ K/k
2
m3/s) m3/s) W)
0 80 60 6 1 0.04 1 7 0.0067 1 86.4 8.73 8.59 0.000
80 60 59 0.04 1 7 0.0064 1 86 .4 8.41
80 60 60 0.04 1 7 0.0066 1 86.4 8.62
0.5 80 60 6 1 0.041 7 0.0066 1 86.4 8.62 8.55 0.000
80 60 63 0.041 7 0.0067 1 86.4 8.73
80 60 59 0.041 7 0.0063 1 86.4 8.30
1 3 80 60 58 0.041 7 0.0063 1 86.4 8.30 8.44 0.002
80 60 62 0.04 1 7 0.0066 1 86.4 8.62
80 60 6 1 0.041 7 0.0064 1 86.4 8.41
1 .5 80 60 62 0.041 7 0.0062 1 86.4 8.08 7.82 0.01 1
80 60 60 0.041 7 0.0059 1 86.4 7.75
80 60 59 0.041 7 0.0058 1 86.4 7.64
2 80 60 59 0.041 7 0.0055 1 86.4 7.21 7.42 0.0 1 8
80 60 59 0.041 7 0.0055 1 86.4 7.21
80 60 62.5 0.041 7 0.0060 1 86.4 7.86
2.5 80 60 59 0.041 7 0.0055 1 86.4 7.21 7.08 0.025
80 60 59 0.041 7 0.0054 1 86.4 7.05
80 60 58 0.04-1 7 0.0053 1 86.4 6.99
3 80 60 58 0.041 7 0.0053 1 86.4 6.92 7. 1 8 0.023
80 60 6 1 0.041 7 0.0056 1 86.4 7.27
80 60 62 0.04 1 7 0.0056 1 86.4 7.36
3.5 80 60 62 0.041 7 0.0056 1 86.4 7.31 7.05 0.025
80 60 59 0.041 7 0.0053 1 86.4 6.92
80 60 59 0.041 7 0.0053 1 86.4 6.92
4 80 60 58 0.041 7 0.005 1 1 86.4 6.66 6.70 0.033
80 60 58 0.041 7 0.0050 1 86.4 6.55
80 60 60 0.041 7 0.0053 1 86.4 6.88
4.5 80 60 62 0.04 1 7 0.0053 1 86.4 6.99 6.73 0.032
80 60 60 0.041 7 0.0051 1 86.4 6.66
80 60 59 0.041 7 0.0050 1 86.4 6.55
5 80 60 60 0.04 1 7 0.0052 1 86.4 6.77 6.67 0. 034
80 60 6 1 0.041 7 0.0052 1 86.4 6.83
80 60 59 0.041 7 0.0049 1 86.4 6.40
5.5 80 60 6 1 0.041 7 0.0053 1 86.4 6.90 6.85 0.030
80 60 59 0.041 7 0.0050 1 86.4 6.55
80 60 62 0.041 7 0.0054 1 86.4 7. 1 0
6 80 60 62 0.041 7 0.005 3 1 86.4 6.88 6.69 0.033
Appendix VII 246
80 60 60 0.041 7 0.0051 1 86 .4 6.64

80 60 60 0.041 7 0.0050 1 86.4 6.55
6.5 80 60 60 0.04 1 7 0.0049 1 86.4 6.40 6.38 0.040
80 60 59 0.04 1 7 0.0049 1 86.4 6.38
80 60 59 0.041 7 0.0049 1 86.4 6.38
7 80 60 60 0.041 7 0.0049 1 86.4 6.38 6.42 0.039
80 60 59 0.04 1 7 0.0048 1 86.4 6.33
80 60 60.5 0.04 1 7 0.0050 1 86.4 6.55
Time ST ET FT FF W exo RS H eal Ave-H eal Rf

2 2 '
(hr) ( OC ) (OC ) (OC ) (x1 0·3 (x1 0·3 (rad/ )
s (kW/m K) (kW/m K) (m K/k
m3/s) m3/s) W)
0 90 70 68 0.04 1 7 0.0083 1 86.4 1 0.70 1 0.77 0.000
90 70 70 0.041 7 0.0083 1 86 .4 1 0.81
90 70 71 0.041 7 0.0083 1 86 .4 1 0.81
0.5 90 70 72 0.041 7 0.0077 1 86.4 9.94 9.65 0.007
90 70 71 0.04 1 7 0.0075 1 86 .4 9.73
90 70 69 0.04 1 7 0.0072 1 86.4 9.29
1 90 70 70 0.041 7 0.0068 1 86 A 8.75 8.61 0.021
90 70 68 0.041 7 0.0064 1 86 .4 8.32
90 70 71 0.04 1 7 0.0068 1 86.4 8.75
1 .5 90 70 71 0.04 1 7 0.0062 1 86.4 8.00 7.71 0.032
90 70 68 0.041 7 0.0058 1 86.4 7.56
90 70 68 0.041 7 0.0058 1 86.4 7.56
2 90 70 70 0.0�n 7 0.0056 1 86 .4 7.24 7. 1 0 0.045
90 70 68 0.041 7 0.0053 1 86 .4 6.92
90 70 70 0.04 1 7 0.0055 1 86 .4 7. 1 3
2.5 90 70 72 0.04 1 7 0.0056 1 86.4 7.24 6.97 0.045
90 70 69 0.04 1 7 0.0053 1 86.4 6.92
90 70 68 0.04 1 7 0.0052 1 86.4 6.76
3 90 70 71 0.04 1 7 0.0055 1 86.4 7. 1 3 7.02 0.047
90 70 70 0.04 1 7 0.0054 1 86.4 7.02
90 70 69 0.04 1 7 0.0053 1 86.4 6.92
3.5 90 70 72 0.041 7 0.0056 1 86.4 7.24 6.95 0.045
90 70 69 0.041 7 0.0053 1 86.4 6.81
90 70 69 0.04 1 7 0.0053 1 86.4 6.81
4 90 70 68 0.04 1 7 0.0052 1 86 .4 6.70 7.05 0.056
90 70 72 0.04 1 7 0.0056 1 86.4 7.24
90 70 72 0.04 1 7 0.0056 1 86.4 7.20
4.5 90 70 72 0.041 7 0.0056 1 86 .4 7.28 7.00 0. 044
90 70 70 0.041 7 0.0054 1 86.4 7.02
90 70 68 0.041 7 0.0052 1 86.4 6.70
Appendix VII 247
5 90 70 71 0.04 1 7 0.0055 1 86.4 7. 1 3 7. 1 7 0.047

90 70 72 0.04 1 7 0.0058 1 86.4 7.4 6
90 70 70 0.04 1 7 0.0053 1 86.4 6.92
5.5 90 70 70 0.04 1 7 0.0057 1 86.4 7.35 7.30 0.043
90 70 71 0.04 1 7 0.0057 1 86.4 7.3 5
90 70 70 0.04 1 7 0.0056 1 86.4 7.20
6 90 70 69 0.04 1 7 0.0055 1 86.4 7. 1 3 7.31 0.047
90 70 70 0.04 1 7 0.0056 1 86.4 7.24
90 70 73 0.04 1 7 0.0058 1 86.4 7.56
Time ST ET FT FF Wexo RS H eal Ave-Heal R,

(hr) (OC ) (OC ) (OC ) (x1 0·3 (x1 0·3 (rad/s) (kW/m2K) (kW/m2 K) (m 2 K1k
m3/s) m3/s) W)
0 90 70 70 0.04 1 7 0.0082 1 04.7 1 0.59 1 0.59 0.000
90 70 70 0.04 1 7 0.0083 1 04.7 1 0.70
90 70 70 0.04 1 7 0.0081 1 04.7 1 0.48
0.5 90 70 69.8 0.04 1 7 0.0065 1 04.7 8.43 8.43 0.024
90 70 68.5 0.04 1 7 0.0063 1 04.7 8.21
90 70 69.9 0.04 1 7 0.0067 1 04.7 8.64
1 90 70 70.5 0.04 1 7 0.0063 1 04.7 8.21 8.07 0.027
90 70 69 0.04 1 7 0.0059 1 04.7 7.67
90 70 72 0.04 1 7 0.0064 1 04.7 8 .32
1 .5 90 70 70.5 0.04 1 7 0.0057 1 04.7 7.35 7.58 0.042
90 70 72 0.04 1 7 0.0060 1 04.7 7.78
90 70 72 0.04 1 7 0.0059 1 04.7 7.61
2 90 70 73 0.04 1 7 0.0058 1 04.7 7.56 7.24 0.038
90 70 70.5 0.041 7 0.0056 1 04.7 7.24
90 70 69 0.04 1 7 0.0053 1 04.7 6.92
2.5 90 70 68 0.04 1 7 0.0052 1 04.7 6.70 6.84 0.055
90 70 69 0.04 1 7 0.0052 1 04.7 6.70
90 70 70 0.041 7 0.0055 1 04.7 7. 1 3
3 90 70 70 0.041 7 0.0053 1 04.7 6.92 7. 1 3 0.050
90 70 72 0.04 1 7 0.0057 1 04.7 7.35
90 70 71 0.04 1 7 0.0055 1 04.7 7. 1 3
3.5 90 70 71 0.041 7 0.0054 1 04.7 7.02 6.81 0.048
90 70 68 0.04 1 7 0.0052 1 04.7 6.70
90 70 68 0.041 7 0.0052 1 04.7 6.70
4 90 70 70 0.04 1 7 0.0053 1 04.7 6.92 7. 1 3 0.050
90 70 71 .5 0.041 7 0.0055 1 04.7 7. 1 3
90 70 72 0.04 1 7 0.0057 1 04.7 7.35
4.5 90 70 69 0.041 7 0.0052 1 04.7 6.70 6.77 0.055
90 70 68 0.041 7 0.0050 1 04.7 6.48
Appendix VII 248
90 70 71 0.041 7 0.0055 1 04.7 7. 1 3

5 90 70 69 0.04 1 7 0.005 1 1 04.7 6.59 6.59 0.057
90 70 68 0.041 7 0.0050 1 04.7 6 .48
90 70 69 0.04 1 7 0.0052 1 04.7 6.70
5.5 90 70 71 0.04 1 7 0.0055 1 04.7 7. 1 3 6.99 0.046
90 70 72 0.04 1 7 0.0055 1 04.7 7. 1 3
90 70 70 0.04 1 7 0.0052 1 04.7 6.70
6 90 70 70 0.041 7 0.0055 1 04.7 7. 1 3 6.84 0.046
90 70 70 0.04 1 7 0.0052 1 04.7 6.70
90 70 70 0.04 1 7 0.0052 1 04.7 6.70
Time 8T ET FT FF WexrJ R8 H eal Ave-Heal R,

(hr) (OC ) (OC ) (OC ) (x1 0·3 ( x1 0·3 (rad/s) (kW/m' K) (kW/m2K) (m2K/k
m3/s) m3/s) W)
0 80 70 70 0.01 67 0.0037 1 86.4 9.51 9.58 0.000
80 70 70 0.01 67 0.0037 1 86.4 9.51
80 70 73 0.01 67 0.0038 1 86.4 9.73
0.5 80 70 72 0.01 67 0.0037 1 86.4 9.51 9.29 0.003
80 70 69 0.01 67 0.0036 1 86.4 9.29
80 70 69 0.01 67 0.0035 1 86.4 9 . 08
1 80 70 70 0.01 67 0.0035 1 86.4 9.08 9. 1 5 0.005
80 70 72 0.01 67 0.0036 1 86.4 9.29
80 70 71 0.01 67 0.0035 1 86.4 9 .08
1 .5 80 70 70 0.01 67 0.0033 1 86.4 8 . 64 8.69 0.01 1
80 70 73 0.01 67 0.0034 1 86.4 8.86
80 70 72 0.01 67 0.0033 1 86.4 8.56
2 80 70 68 0.01 67 0.0030 1 86.4 7.65 7.71 0.025
80 70 71 0.01 67 0.003 1 1 86.4 7.9 1
80 70 68 0.01 67 0.0029 1 86.4 7.56
2.5 80 70 70 0.0 1 67 0.0032 1 86.4 8.21 7.88 0.023
80 70 70 0.01 67 0.0030 1 86.4 7.65
80 70 68 0.01 67 0.0030 1 86.4 7.78
3 80 70 71 0.01 67 0.0030 1 86.4 7.78 7.71 0. 025
80 70 70 0.01 67 0.0030 1 86.4 7.78
80 70 68 0.01 67 0.0029 1 86.4 7.56
3.5 80 70 68 0.01 67 0.0029 1 86.4 7.56 7.78 0 .024
80 70 70 0.01 67 0.0030 1 86.4 7.78
80 70 72 0.01 67 0.003 1 1 86.4 8.00
4 80 70 70 0.0 1 67 0.0031 1 86.4 8 . 08 7.87 0. 023
80 70 71 0.0 1 67 0.0030 1 86.4 7.87
80 70 68 0.01 67 0.0030 1 86.4 7.65
4.5 80 70 70.5 0.01 67 0.0030 1 86. 4 7.78 7.58 0.028
Appendix VII 249
80 70 68 0.0 1 67 0.0029 1 86.4 7.61

80 70 68 0.01 67 0.0028 1 86.4 7.35
5 80 70 70 0.01 67 0.0031 1 86.4 8.00 7.78 0.024
80 70 67 0.01 67 0.0029 1 86.4 7.56
80 70 68 0.01 67 0.0030 1 86.4 7.78
5.5 80 70 70_ 0.01 67 0.0030 1 86.4 7.65 7.66 0.026
80 70 68 0.01 67 0.0029 1 86.4 7.56
80 70 70 0.01 67 0.0030 1 86.4 7.78
6 80 70 70 0.01 67 0.0030 1 86.4 7.65 7.66 0.026
80 70 68 0.01 67 0.0029 1 86.4 7.56
80 70 70 0.0 1 67 0.0030 1 86.4 7.78
Time ST ET FT FF Wem RS Heal Ave- H eal Rf

( ( ( 3 2 � '
O O
(hr) C ) C ) C ) O
(x1 0- (x1 0-3 (rad/s) (kW/m K) (kW/m K) (m Klk
3
m /s) m3/s) W)
0 80 65 67 0.0283 0.0058 1 46.6 1 0. 1 4 9.75 0.000
80 65 65 0.0283 0.0056 1 46.6 9.64
80 65 63.5 0.0283 0.0055 1 46.6 9.47
0.5 80 65 67 0.0283 0.0054 1 46.6 9.41 9.27 0.005
80 65 67 0.0283 0.0054 1 46.6 9.41
80 65 63 0.0283 0.0052 1 46.6 8.98
1 80 65 65 0.0283 0.0050 1 46.6 8.69 8.62 0.0 1 3
80 65 65.5 0.0283 0.005 1 1 46.6 8.78
80 65 63 0.0283 0.0048 1 46.6 8.40
1 .5 80 65 64 0.02-83 0.0045 1 46.6 7.8 2 7.79 0.026
80 65 63 0.0283 0.0043 1 46.6 7.53
80 65 64.5 0.0283 0.0046 1 46.6 8.02
2 80 65 65 0.0283 0.0045 1 46.6 7.8 2 7.70 0.027
80 65 66 0.0283 0.0044 1 46.6 7.6 8
80 65 65 0.0283 0.0044 1 46.6 7.62
2.5 80 65 63 0.0283 0.0040 1 46.6 6.95 7.22 0.036
80 65 64 0.0283 . 0. 0042 1 46.6 7.24
80 65 65 0.0283 0.0043 1 46.6 7.47
3 80 65 65 0.0283 0.0042 1 46.6 7.24 7.29 0.035
80 65 65.5 0.0283 0.0043 1 46.6 7.39
80 65 65 0.0283 0.0042 1 46.6 7.24
3.5 80 65 65 0.0283 0.0042 1 46.6 7.24 7. 1 0 0.038
80 65 63 0.0283 0.0039 1 46.6 6.81
80 65 65 0.0283 0.0042 1 46.6 7.24
4 80 65 65 0.0283 0.0042 1 46.6 7.24 7.24 0.036
80 65 65 0.0283 0.0042 1 46.6 7.24
80 65 65 0.0283 0.0042 1 46.6 7.24
Appendix VII 250
Time ST ET FT FF Wexp RS H eal Ave -H eal Rf

( ( (
(hr) O C ) OC ) OC ) (x1 0-3 (x1 0-3 (rad/s) (kW/m2 K) (kW/m2K) (m2K1k
m3/s) m3/s) W)
0 80 60 60 0.041 7 0.0066 1 04.7 8.62 8 .57 0.000
80 60 60 0.041 7 0.0066 1 04.7 8.62
80 60 60 0.04 1 7 0.0065 1 04.7 8.47
0.5 80 60 62 0.041 7 0.0066 1 04.7 8.69 8.54 0.000
80 60 62 0.041 7 0.0066 1 04.7 8.62
80 60 60 0.04 1 7 0.0063 1 04.7 8.30
1 80 60 60 0.04 1 7 0.0063 1 04.7 8.30 8. 1 9 0.005
80 60 60 0.041 7 0.0063 1 04.7 8.30
80 60 59 0.04 1 7 0.0061 1 04.7 7.97
1 .5 80 60 60 0.04 1 7 0.0058 1 04.7 7.64 7.63 0.0 1 4
80 60 59 0.04 1 7 0.0057 1 04.7 7.42
80 60 60 0.04 1 7 0.0060 1 04.7 7.82
2 80 60 60 0.04 1 7 0.0057 1 04.7 7.47 7.23 0.022
80 60 58 0.04 1 7 0.0054 1 04.7 7. 1 0
80 60 57.5 0.041 7 0.0055 1 04.7 7. 1 4
2.5 80 60 59.5 0.04 1 7 0.0054 1 04.7 7.05 7. 1 5 0.023
80 60 6 1 0.041 7 0.0055 1 04.7 7.21
80 60 6 1 0.041 7 0.0055 1 04.7 7.2 1
3 80 60 60 0.041 7 0.0053 1 04.7 6.88 6.94 0.028
80 60 6 1 0.04 1 7 0.0053 1 04.7 6.99
80 60 6 1 0.041 7 0.0053 1 04.7 6.94
3.5 80 60 57 0.04 1 7 0.0048 1 04.7 6.33 6.49 0.037
80 60 60 0.04 1 7 0.0050 1 04.7 6.48
80 60 6 1 .5 0.041 7 0.005 1 1 04.7 6.66
4 80 60 60 0.041 7 0.0049 1 04.7 6.44 6.24 0.044
80 60 60.5 0.041 7 0.0048 1 04.7 6.33
80 60 58.5 0.04 1 7 0.0046 1 04.7 5.96
. Fresh whey changed
4.5 80 60 60 0.041 7 0.0048 1 04.7 6.29 6.32 0.042
80 60 60.5 0.041 7 0.0048 1 04.7 6.33
80 60 6 1 0.04 1 7 0.0048 1 04.7 6.33
5 80 60 59 0.041 7 0.0046 1 04.7 6.00 6.04 0.049
80 60 60.5 0.041 7 0.0047 1 04.7 6. 1 1
80 60 60 0.041 7 0.0046 1 04.7 6.00
5.5 80 60 60 0.041 7 0.0043 1 04.7 5.63 5.59 0.062
80 60 58 0.04 1 7 0.0042 1 04.7 5.46
80 60 60 0.041 7 0.0043 1 04.7 5.68
6 80 60 60 0.04 1 7 0. 0044 1 04.7 5.79 5.64 0.061
80 60 6 1 0.04 1 7 0.0043 1 04.7 5.68
80 60 59 0.041 7 0.0042 1 04.7 5.46
Appendix VII 25 1
6.5 80 60 59 0.041 7 0.0041 1 04.7 5.35 5.46 0.067

80 60 60.5 0.041 7 0. 0042 1 04.7 5 .46
80 60 61 0.04 1 7 0.0043 1 04.7 5.57
7 80 60 64 0.041 7 0.0043 1 04.7 5.63 5.48 0.066
80 60 62 0.041 7 0.0042 1 04.7 5.46
80 60 61 0.041 7 0.0041 1 04.7 5.35
Fre s h whev changed again
7.5 80 60 60 0.041 7 0.0040 1 04.7 5.24 5 . 09 0.080
80 60 60 0.04 1 7 0.0039 1 04.7 5. 1 3
80 60 58 0.041 7 0.0038 1 04.7 4.91
8 80 60 60 0.04 1 7 0.0038 1 04.7 4.91 4.91 0.087
80 60 58 0.04 1 7 0.0038 1 04.7 4.91
80 60 59 0.04 1 7 0.0038 1 04.7 4.91
8.5 80 60 62 0.041 7 0.0038 1 04.7 4.91 4.81 0.091
80 60 60.5 0.04 1 7 0.0037 1 04.7 4.80
80 60 59 0.041 7 0.0036 1 04.7 4.72
9 80 60 58 0.041 7 0.0035 1 04.7 4.59 4.72 0.095
80 60 60 0.041 7 0.0036 1 04.7 4.76
80 60 61 0.041 7 0.0037 1 04.7 4.80
Time ST ET FT FF Wem RS H eal Ave-H eal Rf

( ( (
(hr) O C ) O C ) OC ) (x1 0-3 (x1 0-3 (rad/s) (kW/m K) (kW/m2K) (m 2K1k
2
m3/s) m3/s) W)
0 70 60 58 0.01 67 0.0032 1 04.7 8.38 8.69 -0.002
70 60 60.5 0.01 67 0.0034 1 04.7 8.95
70 60 60 0.0 1 67 0.0033 1 04.7 8.73
0.5 70 60 60 0.01 67 0.0033 1 04.7 8.73 8 . 63 -0.001
70 60 59 0.01 67 0.0032 1 04.7 8.43
70 60 60 0.0 1 67 0.0033 1 04.7 8.73
1 70 ' 60 60 0.01 67 0.0033 1 04.7 8.52 8 . 66 -0.001
70 60 60 0.01 67 0.0033 1 04.7 8.52
70 60 6 1 .5 0.0 1 67 0.0034 1 04.7 8.95
1 .5 70 60 60 0.0 1 67 0.0033 1 04.7 8.52 8.52 0.002
70 60 59 0.0 1 67 0.0033 1 04.7 8.52
70 60 58 0.01 67 0.0033 1 04.7 8.52
2 70 60 62 0. 0 1 67 0.0033 1 04.7 8.52 8.30 0. 004
70 60 60 0.01 67 0.0032 1 04.7 8.30
70 60 59 0.0 1 67 0.0031 1 04.7 8.08
2.5 70 60 59.5 0.0 1 67 0.0032 1 04.7 8.30 8.30 0.004
70 60 60 0.01 67 0.0032 1 04.7 8.30
70 60 61 0.0 1 67 0.0032 1 04.7 8.30
3 70 60 6 1 .5 0.01 67 0.0033 1 04.7 8.52 8. 1 5 0.006
Appendix VII 252
70 60 59 0.01 67 0.003 1 1 04.7 8.08

70 60 57 0 . 0 1 67 0.0030 1 04.7 7.86
3.5 70 60 64 0.01 67 0.0032 1 04.7 8.43 8.34 0.003
70 60 6 1 0.01 67 0.0032 1 04.7 8.30
70 60 60 0.01 67 0.0032 1 04.7 8.30
4 70 60 63 0.01 67 0.0032 1 04.7 8.47 8.28 0.004
70 60 60.5 0.0 1 67 0.0032 1 04.7 8.30
70 60 60 0.0 1 67 0.0031 1 04.7 8.08
4.5 70 60 64 0.01 67 0.0033 1 04.7 8.52 8.35 0.003
70 60 60 0.01 67 0.0032 1 04.7 8 .25
70 60 61 0.01 67 0.0032 1 04.7 8.30
5 70 60 60 0.0 1 67 0. 0032 1 04.7 8.30 8.22 0.005
70 60 61 .5 0.01 67 0.0032 1 04.7 8.30
70 60 60 0.01 67 0.003 1 1 04.7 8.08
5.5 70 60 60 0.01 67 0.0032 1 04.7 8 .30 8. 1 5 0.002
70 60 59 0.01 67 0.003 1 1 04.7 8.08
70 60 59 0.01 67 0.0031 1 04.7 8.08
6 70 60 60 0.01 67 0.0032 1 04.7 8.30 8. 1 5 0.006
70 60 59 0.01 67 0.003 1 1 04.7 8 .08
70 60 59 0.01 67 0.0031 1 04.7 8.08
Time ST ET FT FF W eX!) RS H eal Ave-Heal R,

2 2 2
(hr) (OC ) (OC ) (OC ) (x1 0-3 (x1 0 -3 (rad/s) (kW/m K) (kW/m K) (m K1k
m3/s) m3/s) W)
0 70 60 60 0.04 1 7 0.0036 1 86.4 9.48 9.78 0.000
70 60 62 0.04 1 7 0.0038 1 86.4 9.83
70 60 63 0.04 1 7 0.0038 1 86.4 1 0.04
0.5 70 60 60 0.04 1 7 0.0033 1 86.4 8.73 8.85 0.01 1
70 60 61 0.04 1 7 0.0035 1 86.4 9.08
70 60 60 0.04 1 7 0.0033 1 86.4 8.73
1 70 60 60.5 0.04 1 7 0.0034 1 86.4 8.95 8.52 0.0 1 5
70 60 57.5 0.04 1 7 0.0030 1 86.4 7.86
70 60 60 0.04 1 7 0.0033 1 86.4 8.73
1 .5 70 60 60.5 0.04 1 7 0.0032 1 86.4 8.47 8.50 0 .0 1 5
70 60 58.5 0.04 1 7 0.0031 1 86.4 8.08
70 60 61 0.04 1 7 0. 0034 1 86.4 8.95
2 70 60 60 0.04 1 7 0.0034 1 86.4 8.95 8.68 0.01 3
70 60 60 0.04 1 7 0.0033 1 86.4 8.56
70 60 60 0.04 1 7 0.0033 1 86.4 8.52
2.5 70 60 60 0.04 1 7 0.0033 1 86.4 8.65 8.65 0.01 3
70 60 60 0.04 1 7 0.0033 1 86.4 8 . 65
70 60 60 0.04 1 7 0.0033 1 86.4 8.65
Appendix VII 253
3 70 60 59 0.041 7 0.0032 1 86.4 8.30 8.66 0.0 1 3

70 60 60 0.04 1 7 0.0033 1 86.4 8.52
70 60 62 0.041 7 0.0035 1 86.4 9. 1 7
Time 5T ET FT FF W exo R5 H eal Ave-Heal R,

( ( (
(hr) O C ) OC ) O C ) (x1 0·3 (x1 0·3 2
(rad/s) (kW/m K) (kW/m2K) 2
(m K/k
m3/s) 3
m /s) W)
0 80 70 71 0.041 7 0 .0038 1 86.4 9.73 9.48 0.001
80 70 7 1 .5 0.041 7 0.0037 1 86.4 9.64
80 70 68.5 0.041 7 0.0035 1 86.4 9.08
0.5 80 70 70 0.041 7 0.0032 1 86.4 8.21 8.33 0.0 1 6
80 70 7 1 .5 0.041 7 0.0034 1 86.4 8.77
80 70 69.8 0.041 7 0.0031 1 86.4 8.00
1 80 70 67.5 0.041 7 0.0028 1 86.4 7.35 7.92 0.022
80 70 71 0.041 7 0.003 1 1 86.4 8.00
80 70 72.5 0.041 7 0.0033 1 86.4 8.43
1 .5 80 70 70 0.041 7 0.0030 1 86.4 7.78 7.78 0 . 024
80 70 70 0.04 1 7 0.0030 1 86.4 7.78
80 70 70 0.04 1 7 0.0030 1 86.4 7.78
2 80 70 72 0.04 1 7 0.0031 1 86.4 7.91 7. 1 5 · 0.036
80 70 70 0.041 7 0. 0027 1 86.4 6.92
80 70 68 0.041 7 0.0026 1 86.4 6.61
2.5 80 70 70 0.041 7 0.0027 1 86.4 6.87 6.97 0 . 039
80 70 72.5 0.041 7 0.0029 1 86.4 7.56 .
80 70 67 0.041 7 0.0025 1 86.4 6.48
3 80 70 70 0.041 7 0.0027 1 86.4 6.92 6.92 0.040
80 70 70 0.041 7 0.0027 1 86.4 6.92
80 70 70 0.041 7 0.0027 1 86.4 6.92
3.5 80 70 66 0.041 7 0.0023 1 86.4 6.05 6.60 0. 047
80 70 69 0.04 1 7 0.0026 1 86.4 6.83
80 70 70 0.041 7 0.0027 1 86.4 6.92
Time 5T ET FT FF W exo R5 H eal Ave-H eal R,

( ( (
(hr) O C ) OC ) O C ) (x1 0·3 (x1 0·3 (rad/s) (kW/m K) (kW/ m 'K) (m ' Klk
2
m3/s) m3/s) W)
0 90 70 70 0.01 67 0.0072 1 04.7 9.29 9.29 0.000
90 70 70 0.01 67 0.0072 1 04.7 9.29
90 70 70 0.01 67 0.0072 1 04.7 9.29
1 90 70 70 0.01 67 0.006 1 1 04.7 7.91 7.72 0 . 032
90 70 70 0.01 67 0.0059 1 04.7 7.67
90 70 70 0.01 67 0 .0058 1 04.7 7.56
Appendix VII 254
2 90 70 70 0.0 1 67 0.005 1 1 04.7 6.66 6.63 0.056

90 70 70 0.0 1 67 0.0051 1 04.7 6.59
90 70 72 0.01 67 0.0051 1 04.7 6.66
3 90 70 70 0.0 1 67 0.0047 1 04.7 6. 1 2 6.04 0.069
90 70 70 0.0 1 67 0 .0046 1 04.7 5.94
90 70 70 0.0 1 67 0 .0047 1 04.7 6.05
4 90 70 70 0.0 1 67 0.0047 1 04.7 6.05 6.05 0.07 1
90 70 70 0.01 67 0 .0047 1 04.7 6.05
90 70 70 0.01 67 0 .0047 1 04.7 6.05
5 90 70 70 0.01 67 0.0047 1 04.7 6 .05 6.05 0.071
90 70 70 0.01 67 0 .0047 1 04.7 6.05
90 70 70 0.01 67 0.0047 1 04.7 6.05
Ti me ST ET FT FF W an RS H eal Ave-Heal RI
( ( (
O O
(hr) C ) C ) C ) O
(x1 0·3 (x1 0·3
2
(rad/s) (kW/m K) (kW/m 2K) (m 2K1k
m3/s) m 3/s) W)
0 80 60 60 0.01 67 0.0068 1 04.7 8.95 8.95 - 0.005
80 60 60 0.01 67 0.0068 1 04.7 8.95
80 60 60 0.01 67 0.0068 1 04.7 8.95
1 80 60 60 0.01 67 0.0067 1 04.7 8.73 8.72 -0.002
80 60 60 0.01 67 0 . 0066 1 04.7 8.69
80 60 60 0.01 67 0.0067 1 04.7 8.73
2 80 60 60 0.01 67 0.0064 1 04.7 8.41 8.41 0.002
80 60 60 0.01.67 0.0065 1 04.7 8.52
80 60 60 0.0 1 67 0.0063 1 04.7 8.30
3 80 60 60 0.0 1 67 0.0061 1 04.7 7.97 7.97 0.0 1 4
80 60 60 0.01 67 0.0061 1 04.7 7.97
80 60 60 0.01 67 0.0061 1 04.7 7.97
4 80 . 60 60 0.01 67 0.0060 1 04.7 7.82 7.77 0.0 1 2
80 60 60 0.01 67 0. 0059 1 04.7 7.75
80 60 60 0.01 67 0 .0059 1 04.7 7.75
5 80 60 60 0.01 67 0.0057 1 04.7 7.42 7.50 0.0 1 7
80 60 60 0.01 67 0 . 0058 1 04.7 7.53
80 60 60 0.01 67 0.0058 1 04.7 7.53
Appendix VII 255
Time ST ET FT FF W exp RS Heal Ave-Heal

( ( ( 2
(hr) CO
C C O O
(X1 0'3 (x 1 0-3 (rad/s) (kW/m K) (kW/m2K)
) ) ) m3
/s) m3/s)
5 70 65 66 0.0283 0.00 1 3 1 46.6 6.69 6.37
70 65 65.5' 0.0283 0.00 1 2 1 46.6 6.26
70 65 64.5 0.0283 0.00 1 2 1 46.6 6. 1 7
10 75 65 65 0.0283 0.0026 1 46.6 6.73 6.66
75 65 64 0.0283 0.0025 1 46.6 6.52
75 65 65 0.0283 0.0026 1 46.6 6.73
20 85 65 65.5 0.0283 0.0056 1 46.6 7.34 7.37
85 65 65.5 0.0283 0.0057 1 46.6 7.39
85 65 65 0.0283 0.0057 1 46.6 7.39
25 90 65 65 0.0283 0.0072 1 46.6 7.52 7.63
90 65 65 0.0283 0.0073 1 46.6 7.65
90 65 65 0.0283 0.0074 1 46.6 7.73
30 95 65 66 0.0283 0.0090 1 46.6 7.82 7.84
95 65 65 0.0283 0.0089 1 46.6 7.75
95 65 65 0.0283 0.0092 1 46.6 7.96
Heat Transfer and Fouling i n F i l m Evapo rator with Rotating S u rfac es 256
Appendix VITI
A typical native-PAGE patterns of whey proteins
;.
� .,
...,....,. . �
� �.
�
I . �.:...-��'�� ';'-"�( .. -
1 ....... - - - - - - ·'.· · BSA
•
I .........
_ _
. ... . ':'L' �; .;. . .,. " ",1

!I � . >..�.
� . ;
•
' .
� ,0
I
a.-Ia
t:= . 13-lg A
. ;� .. 13-lg B
.,
...
Time (hr): o 0 1 2 3 4 5 6 6

Heat Transfer Rotating Surfaces

Uploaded by

Copyright:

Available Formats

Heat Transfer Rotating Surfaces

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Heat Transfer Rotating Surfaces

Uploaded by

Copyright:

Available Formats

Copyright is owned by the Author of the thesis.

Permission is given for

HEAT TRANSFER AND FOULING

A THESIS PRESENTED IN PARTIAL FULFILMENT OF

THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY IN FOOD TECHNOLOGY

B.E., M.Tech. (Honours)

Heat Transfer and Fouling in Film Evaporators with Rotating Surfaces ii

By using Nusselt-type assumptions, a theoretical model, which was used to

The experimental studies on heat transfer were conducted on a Centritherm

With the Centritherm evaporator, good agreement between theoretical and

temperature and the temperature difference between steam condensing and

liquid evaporating temperatures. In general, the higher the evaporating

Albumins (BSA) denatured after running the evaporator, at a evaporating

temperature of 70°C and a temperature difference of 20K, for 6 hours. Though

attached to the surface. By association with other depositable materials existing

coefficients significantly. This was attributed to the lower thermal conductivity

velocity. This effect was more significant at lower evaporation temperatures.

reduce the rate of fouling.

depositable materials are much more easily adsorbed on fouled or unclean

solutions were introduced suggested that the concentration of activated

molecules in the solutions strongly affected the fouling process. A possible

mechanism of whey fouling on the rotating surface was proposed.

in which a vapour compressor would be integrated with the rotating surface.

the on-farm evaporation system, which requires an evaporator with high

efficiency, compact, and minimum heat load to milk, it is suggested that a

evaporator) would be optimum and worth to explore.

I wish to express my deepest appreciation and gratitude to my supervisor, Mr.

friendship, and to my co-supervisor, Dr. Osvaldo Campanella, for his

suggestions, valuable advice and encouragement throughout this study at

the duration of my study.

their generous help, continuous care and friendship.

my friends, fellows at Massey University, and the members of New Zealand­

China Friendship Society, who in many ways contributed to the completion of

this work. Especially, I have to mention that when I was involved in an

through a difficult time. Their invaluable assistance is very much appreciated

and will be always remembered.

I also thank my father, mother, sister, grandmother, parents-in-law as well as the

ABSTRACT .... ............ ...... ........ ...... . . .......... ii

TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii

LIST OF FIGURES ................................ .......... xv

LIST OF TABLES .. ................ .............. .. .... ..... xix

LIST OF PUBLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxiv

PART I GENERAL INTRODUCTION

CHAPTER 1 INTRODUCTION . ... .................... .......... 1

1.1 Evaporation . . ............. ................ .......... 1

1.2 Evaporation technology used in the dairy industry . .......... 7

1.3 The thin film evaporator with rotating heating surface . . . . . . . . 8

1.4 Fouling in evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.5 On-farm evaporation system . . . . . . . . . . . . . . . . . . . . . . . . . .. 10

CHAPTER 2 OBJECTIVES . ........................... ......... 1 3

PART II HEAT TRANSFER IN EVAPORATORS WITH

CHAPTER 3 HEAT TRANSFER IN THIN FILM EVAPORATORS -

LITERATURE REVIEW . .... .............. ........ .. 15

3.1 Liquid film flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3.1.1 The flow regimes of the liquid film . . . . . . . . . . . . . . . 16

3.1.2 Flow features of the falling liquid films . . . . . . . . . . . . 18

3.2 Liquid boiling phenomena . . . . . . . . . . . . . . . . . . . . . . . 19

3.2.1 Pool boiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3.2.2.Flow boiling . . . . . . . . . . . . . . . . . . . . ' ... .......... 2 2

my friends, fellows at Massey University, and the members of New Zealand