close

Вход

Забыли?

вход по аккаунту

?

Continuous demulsification of emulsions using microwaves

код для вставкиСкачать
1+1
National Library
of C a n a d a
Bibliothdque nationalc
du C anada
Acquisitions and
Bibliographic Services Branch
Direction d e s acquisitions et
d e s services bibliographiques
395 Wellington Street
O ttaw a, Ontario
K 1A 0N 4
395, rue Wellington
O ttaw a (Ontario)
K 1A 0N 4
Khi> fife
I *it t''e
Votte
p
Notre
NOTICE
AVIS
The quality of this microform is
heavily dependent upon the
quality of the original thesis
submitted
for
microfilming.
Every effort has been made to
ensure the highest quality of
reproduction possible.
La quality de cette microforme
depend grandement de la quality
de la th&se soumise au
microfilmage. Nous avons tout
fait pour assurer une quality
superieure de reproduction.
If pages are missing, contact the
university which granted the
degree.
S’il manque des pages, veuillez
communiquer avec I‘jmversit6
qui a confere le grade.
Some pages may have indistinct
print especially if the original
pages were typed with a poor
typewriter ribbon or if the
university sent us an inferior
photocopy.
La quality d ’impression de
certaines pages peut kisser a
desirer, surtout si les pages
originates
ont
et6
dactylographies a I’aide d ’un
ruban us6 ou si I’universit6 nous
a fait parvenir une photocopie de
q u a lii inferieure.
Reproduction in full or in part of
this microform is governed by
the Canadian Copyright Act,
R.S.C. 1970, c. C-3G, and
subsequent amendments.
La reproduction, meme partielle,
de cette microforme est soumise
a la Loi canadienne sur le droit
d’auteur, SRC 1970, c. C-30, et
ses amendements subs6quents.
Canada
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
UNIVERSITY OF ALBERTA
CONTINUOUS DEM ULSIFICATIO TOF EMULSIONS USING
MICROWAVES
BY
KEITH WILLIAM REDFORD
A thesis subm itted to the Faculty of G raduate Studies in partial fulfillment of
the requirements for the degree of M aster of Science.
D EPA RTM EN T OF CHEMICAL ENGINEERING
Edmonton, Alberta
Spring 1993
R e p r o d u c e d with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
1*1
National Library
of C a n a d a
Bibliothfeque nationale
du C anada
Acquisitions and
Bibliographic Services Branch
Direction d e s acquisitions et
d e s services bibliographiques
395 Wellington Street
O ttaw a, O ntario
K 1A 0N 4
395, ru e Wellington
O ttaw a (Ontario)
K1A0N4
Y ixii ftk'
i \ j f t>h*
V o tte tQ tfre n c *
Nvttre
rtWriwve
The author has granted an
irrevocable non-exciusive licence
allowing the National Library of
Canada tp reproduce, loan,
distribute or sell copies of
his/her thesis by any means and
in any form or format, making
this thesis available to interested
persons.
L’auteur a accorde une licence
irrevocable et non exclusive
permettant a la Bibliothdque
nationale
du
Canada
do
reproduire, prefer, distribuer ou
vendre des copies de sa thdse
de quelque manidre et sous
quelque forme que ce soit pour
mettre des exemplaires de cette
these a la disposition des
personnes interessees.
The author retains ownership of
the copyright in his/her thesis.
Neither the thesis nor substantial
extracts from it may be printed or
otherwise reproduced without
his/her permission.
L’auteur conserve la propri6t6 du
droit d’auteur qui protege sa
these. Ni la these ni des extraits
substantiels de celle-ci ne
doivent etre
imprimis ou
autrement reproduits sans son
autorisation.
ISBN
0 -3 15 -82211-2
Canada
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
UNIVERSITY OF ALBERTA
RELEASE FORM
NAME OF AUTHOR:
KEITH W. REDFORD
TITLE OF THESIS:
CONTINUOUS DEM ULSIFICATION
EMULSIONS USING MICROW AVES
DEGREE:
OF
M ASTER OF SCIENCE
YEAR THIS DEGREE GRANTED: 1993
Permission is hear by granted to the University of A lberta Library to
reproduce single copies of this thesis and to lend or sell such copies for
private, scholarly or scientific research purposes only.
The author reserves all other publication and other rights in association
with the copyright in the thesis, and except as hereinbefore provided
neither the thesis nor any substantial portion thereof may be printed or
otherwise reproduced in any m aterial form whatever w ithout the
author’s prior w ritten permission
Keith W. Redford
Box 3004, F t. Sask.
Alberta, Canada
T8L 2T1
January 15, 1993
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
UNI V ERS ITY O F A LB ER T A
FACULTY OF GRADUATE STUDIES AND RESEARCH
The undersigned certify th a t they have read, and recommended to the Faculty
of G raduate studies and Research for acceptance, a thesis entitled Continuous
Demulsification of Emulsions using Microwaves subm itted by Keith W. Redford
in partial fulfillment of the requirem ents for the degree of M aster of Science.
H/aA
/
J. H. Masliyah (supervisor)
D. G. Fisher
R N M fy e i
\
& _______________
W. S. Tortike
Janu ary 13, 1993
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
ABSTRACT
Demulsification is an im portant process for the petroleum industry.
Both o il-in -w ater and w a ter-in -o il emulsions occur frequently at various
stages during petroleum production. It has been recently suggested th at
microwaves might be used to affect more efficient demulsification of both types
of petroleum emulsions. Previous studies of microwave demulsification,
however, have employed batch methods. This study considers microwave
demulsification in a continuous flow system.
A continuous microwave demulsification system was designed and built,
using a domestic microwave oven as the microwave source. A m athem atical
therm al model of the microwave system was developed to predict the
tem perature profile of pure fluids and emulsions w ithin the system. The model
was tested using pure fluids, w a ter-in -o il emulsions, and o il-in -w a te r
emulsions. The degree of demulsification achieved by th e microwave system
was investigated. A conventional constant tem perature bath system was also
assembled. The degree of demulsification achieved by this system was
compared to th at achieved using the microwave demulsification system.
It was found th at the developed model predicted the tem perature profiles
for the case where boiling did not occur for all fluid types tested. Tem perature
gradients for the non-boiling case were quite linear.
It was found th a t the emulsion type was critical in determ ining the
degree of demulsification, as expressed by the recovery of the dispersed phase.
For the present experimental setup, the conventional system was found to give
b etter recovery of the dispersed phase th an the microwave system.
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
ACKNOW LEDGEMENTS
I would like to thank Dr. J.H. Masliyah for his guidance throughout this
research. I would also like to thank the Alberta Oil Sands Technology and Research
Authority for Gnancing the project. Thanks is also due to Dr. W.R. Tinga and
Dr. W. Xi, from the Departm ent of Electrical Engineering (University of Alberta).
Finally, I would like to express my gratitude to all the editors, reviewers, and
examiners who gave their tim e to insure the quality of this thesis.
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
TA B LE OF CO NT EN TS
page
Chapter 1 Introduction
Chapter 2
Literature Review
C hapter 3 The Nature of Emulsions
C hapter 4
C hapter 5
C hapter 6
1
4
7
3.1
W hat are Emulsions
7
3.2
Factors th at Stabilize Emulsions
8
3.3
Emulsion Breaking Techniques
10
The Nature u*. licrowaves
12
4.1
W hat are Microwaves
12
4.2
Commercial Use of Microwaves
13
4.3
Microwave Heating
13
4.4
Microwaves and Emulsions
16
4.5
Safety with Microwaves
18
Modelling a Continuous Microwave Heating System
20
5.1
The N on-boiling Condition
20
5.2
The Boiling Condition
24
Model Validation Through Comparison with
Experim ental Data
29
6.1
Experim ental A pparatus
29
6.2
Microwave Heating of W ater and Saline
30
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
C hapter 7
Experiments in Continuous Microwave Demulsification
35
7.1
Experimental Setup
35
7.2
Demulsification of Saline-in-B ayol
Emulsions
36
7.3
Demulsification of B ayol-in-S aline Emulsions
38
7.4
Demulsification of B itum en-in-S aline Emulsions
39
7.5 Conventional Demulsification Equipment
41
7.6 Conventional Demulsification of B itum en-in-S aline
Emulsions
42
7.7 Conclusions
43
Figures
45
Bibliography
87
R ep ro d u ced with p erm ission of the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
lIST-QEJifiU&ES
page
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Particle deposition on an emulsion
droplet
45
Polymer chain interactions between
emulsion droplets resulting in
stabilization of the emulsion:
a) physical blockage of droplet
contact b) osmotic repulsion due
to chain interactions
46
The electric double layer on an
emulsion droplet
47
Movement of charged particles in a
microwave induced electric field
48
Movement of dipolar species in a
microwave induced electric field
49
Tem perature dependence of the
effective dissipation factor for
distilled w ater
50
Frequency dependence of the
effective dissipation factor for
a 0.1 M saline solution
51
T em perature dependence of microwave
penetration depth for distilled
water at various frequencies
52
T em perature dependence of the
dielectric constant for distilled
w ater and 0.1 M saline
53
Frequency dependence of the
dielectric constant for distilled
w ater and 0.1 M saline
54
Element for evaluation of heat and
mass transfer within a coil heated
by microwave irradiation
55
Schematic diagram of continuous
microwave heating apparatus
56
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Figure 13
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18
Figure 19
Loss of tem perature with time for
an insulated coil containing
m icrowave-heated 0.0171 M saline
solution
57
Experimental and model based
tem perature profiles for distilled
water flowing through a continuous
microwave heating system at
0.0016 L/s (measured at the inlet
tem perature)
58
Experimental and model based
tem perature profiles for a
0.0171 M saline solution flowing
through a continuous microwave
heating system at 0.0016 L/s
(measured at the inlet tem perature)
59
Experim ental and model based
tem perature profiles for a 0.06 M
saline solution flowing through a
continuous microwave heating system
at 0.0016 L/s (measured at the
inlet tem perature)
60
Experim ental and model based
tem perature profiles for a 0.12 M
saline solution flowing through a
continuous microwave heating system
at 0.0016 L/s (measured at the
inlet tem perature)
61
Comparison of experim ental and
model tem perature profiles for
distilled w ater and 0.06 M saline
solution flowing through a
continuous microwave heating system
at 0.0016 L /s (measured at the
inlet tem perature)
62
Experim ental and model based
tem perature profiles for a 0.1 M
saline solution flowing through a
continuous microwave heating system
at 0.0029 L /s (measured at the
inlet tem perature)
63
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Figure 20
Figr ;* 21
Figure 22
Figure 23
Figure 24
Figure 25
Figure 26
Figure 27
Experimental and model based
tem perature proGles for distilled
water flowing through a continuous
microwave heating system at
0.0009 L/s (measured at the inlet
tem perature)
64
Experimental and model based
tem perature profiles for a 0.06 M
saline solution flowing through a
continuous microwave heating system
at 0.0009 L/s (measured at the
inlet tem perature)
65
Experim ental and Model based
tem perature profiles for a 0.12 M
saline solution flowing through a
continuous microwave heating system
at 0 0008 L/s (measured at the
inlet tem perature)
66
Comparison of experimental and
model based tem perature profiles
for distilled w ater and 0.06 M
saline solution flowing through a
continuous microwave heating system
at 0.0009 L/s (measured at the
inlet tem perature)
67
Power absorbed by a 0.06 M saline
solution flowing through a
continuous microwave heating system
as a function of flowrate
68
Power absorbed by distilled water
flowing through a continuous
microwave heating system as a
function of flowrate
69
Effective dissipation factors for
w ater and 0.1 M saline at 2800 MHz
as a function of tem perature
70
Layer identification for
demulsified and undemulsified
0.06 M saline in Bayol-35 emulsions
71
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Figure 28
Figure 29
Figure 30
Figure 31
Figure 32
Figure 33
Figure 34
Figure 35
Recovery of Bayol-35 and 0 06 M
saline from 0.06 M saline in
Bayol-35 emulsions at various
dispersed phase concentrations as a
function of coil outlet tem perature
72
Recovery of Bayol-35 and 0 06 M
saline from 0.06 M saline in
Bayol-35 emulsions at various
dispersed phase concentrations as a
function of initial measuring
vessel tem perature
73
Recovery of Bayol-35 and 0.06 M
saline from 0.06 M saline in
Bayol-35 emulsions at various
dispersed phase concentrations as a
function of fluid flowrate
74
Experim ental and model based
tem perature profiles for a 50%
0.06 M saline solution in Bayol-35
emulsion flowing through a
microwave heating system at
0.0018 L/s (measured at the inlet
tem perature)
75
Layer identification for
demulsified and undemulsified
B ayol-35 in 0.06 M saline
solution emulsions
76
Recovery of Bayol-35 and 0.06 M
saline from Bayol-35 in 0.06 M
saline emulsions at various
dispersed phase concentrations as a
function of coil outlet tem perature
77
Recovery of Bayol-35 and saline
from 50% Bayol-35 in saline
emulsions at various saline
concentrations as a function of
coil outlet tem perature
78
Experim ental and model based
tem perature profiles for a 50%
Bayol-35 in 0.06 M saline solution
emulsion flowing through a
microwave heating system at
0.0025 L/s (measured a t the inlet
tem perature)
79
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
Figure 36
Figure 37
Figure 38
Figure 39
Figure 40
Figure 41
Figure 42
Layer identification for
demulsified and undemulsified
bitum en in 0.06 M saline emulsions
80
Recovery of bitumen and 0.06 M
saline from bitumen in 0.06 M
saline emulsions at various
dispersed phase concentrations
as a function of coil outlet
tem perature
81
Experim ental and model based
tem perature profiles for a 14%
Bitumen in 0.06 M saline emulsion
flowing through a continuous
microwave heating system at
0.0019 L/s (measured at the inlet
tem perature)
82
Schematic diagram of a conventional
continuous heating system
83
Comparison of bitumen and 0.06 M
saline recovery for microwave
heated and conventionally heated
emulsions
84
Comparison of experimentally based
tem perature profiles between
microwave heated and conventionally
heated emulsions of 14% bitumen in
0.06 M saline solution flowing at
0.0019 L/s (measured at the inlet
tem perature)
85
Percent w ater remaining in the
emulsion for microwave heated and
conventionally heated 30%
w ater—in-oil(B ayol—35) emulsions
as a function of the water phase
of the partially demulsified
emulsion, (from Pal and Masliyah,
1991)
86
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission .
N O M EN C LA TU R E
Alphanumeric Symbols
heat capacity
kg °C
E
rms
root mean squared electric field strength (V /m )
f
H
K
frequency (Hz)
enthalpy (J/k g )
phase velocity ratio
rti
P
av
R
mass flowrate (kg/s)
average power absorbed (W )
T
tem perature (°C)
T
ac
u
tube radius (m)
oven cavity tem perature (°C)
W
overall heat transfer coefficient
m
v
V
w
x
X
volume of m aterial (m )
weight fraction of the dispersed phase
axial distance along the coil (m)
Lockhart-M artinelli param eter
Greek Symbols
a
X
e
e
e"
A
Lv
P
T
fraction of element occupied by liquid
mass dryness fraction
complex dielectric constant
dielectric constant
dissipation factor
heat of vaporization (J/k g )
density (kg/m )
microwave exposure tim e (s)
volume fraction of the dispersed phase
Subscripts
eff
h
L
m
v
1
2
effective
homogeneous
liquid
emulsion
vapor
suspended phase
continuous phase
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
CH APTER 1
INTRODUCTION
During the production and processing of oil stream s, w ater can become
mixed with the oil. This can be either unintentional, such as when w ater is
pumped up with the oil from an oil field, or intentional, such as in the Clark
Hot W ater Process. The mixing of these two immiscible phases may result in
the formation of an emulsion, wherein one liquid is dispersed as droplets in the
other liquid. W ater and oil emulsions can be of two types, w a ter-in -o il or
o il-in -w ater. A w a ter-in -o il emulsion has w ater droplets suspended in a
'’ontinuous oil phase. A oil—in—water emulsion contains oil droplets suspended
in a continuous w ater phase. Often, w ater—in—oil emulsions must be
demulsified before any upgrading of th at oil can be done. Demulsification of
o il-in -w a te r emulsions m ust often be performed to make w ater reusable or to
comply with environm ental regulations.
Emulsions do not minimize the surface area between their component
phases. This m eanp th a t in the thermodynamic sense, all emulsions are
unstable. There are, however, a number of kinetic factors th a t can increase the
stability of emulsions. This is an undesirable situation. Methods for
destabilizing and coalescing highly stable emulsions are therefore of interest to
the petroleum industry. Present des- -bilizing techniques involve more art th an
science. Heating, chemical addition, and centrifugation are all known to
destabilize emulsions and enhance coalescence.
W hile heating and centrifugation are fairly simple processes, chemical
addition is very complex. The wide variety of oil stream compositions and
1
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
possible chemical treatm ents makes the selection of an appropriate chemical
treatm ent by and large a trial and error process. Inappropriate addition of a
chemical can result in an emulsion th at is more stable than the original, thus
increasing the cost and tim e required to achieve demulsification.
Suggestions th at microwave heating may enhance the demulsification
process (as compared to normal demulsification by heating) have come from
several sources. Making use of batch tests, Wolf, in his 1986 patent, claimed
th a t microwaves could be used to enhance existing separation methods or as a
stand alone process. Fang et al. (1988 1989) presented additional batch studies
on the effectiveness of microwave heating in demulsification. Their work
included field tests on waste oil sludge in a 10 foot diam eter tank. Fang et al
suggested th a t the electrical properties of microwaves enhanced demulsification.
Pal et al. in 1990, investigated emulsions of Bayol—35 and w ater under batch
conditions, using a conventional microwave oven. Pal et al also suggested that
in addition to being an efficient heating method, microwaves have electrical
properties which contribute to demulsification. If a sufficiently high recovery
could be attained with microwave exposure, one might elim inate the need for
th e use of chemicals in the de—emulsification processes altogether. As batch
demulsification d ata cannot be directly used to predict results in a continuous
system, there is a need for microwave tests to be conducted using a bench scale
continuous system. Such continuous microwave heating tests have been
conducted for the pasteurization of milk (K urda et al 1989, Ozilgen and
Ozilgen 1991), these studies are not generally applicable to petroleum and water
systems as no general model was provided.
A detailed review of the m entioned demulsification literature is provided
2
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
in C hapter 2. C hapter 3 provides the relevant theoretical background on
emulsions and on demulsification. In C hapter 4, im portant principles behind
microwaves and their effects on materials are presented, with particular
emphasis on their effects on heterogeneous mixtures. A model to predict the
tem perature profile for single and two—phase flow is presented in C hapter 5.
Comparisons between the model and the experim ental d a ta is given in
C hapter 6. Results for the microwave demulsification of emulsions are
presented in C hapter 7. A comparison between microwave demulsification and
conventional demulsification is also included in C hapter 7, followed by a list of
the prim ary conclusions th at have been made.
3
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
C H A PTER 2
LITERATURE REVIEW
The first document w ritten exclusively about microwave demulsification
of hydrocarbon-in—water and w ater-in-hydrocarbon emulsions was Nicholas 0 .
W olf’s 1986 patent entitled "Use of Microwave Radiation in Separating
Emulsions and Dispersions of Hydrocarbons and W ater". In this work, Wolf
claims invention of a technique for microwave demulsification of both
w a te r-in -o il and o il-in -w a te r emulsions. W olf supports his claim by providing
two examples of the beneficial effects of microwaves on emulsions.
In his first example, 100 ml samples of 30% (by weight) w ater-in -o il
emulsion were placed in a microwave system. Microwave radiation exposure
tim e was for either two 20 second intervals or two 30 second intervals. Two
different surfactants were used. W hen emulsions containing a nonionic
surfactant were irradiated, w ater recoveries of 17% and 50% (by weight) were
noted one hour after exposure. W hen a cationic surfactant was used to create
th e emulsion, w ater recoveries of 23% and 33% (by weight) were achieved one
hour after exposure. Emulsions were kept in a 60°C constant tem perature bath
during the one hour waiting period.
In his second example, a combination of chemicals and microwave
radiation was used to demulsify a "tight" (i.e. well stabilized) 40% (by weight)
o il-in -w a te r emulsion. Recovery of w ater 30 minutes after exposure and with
subsequent chemical addition was found to be 70% (by weight). W ithout the
microwave exposure, but with chemical addition, w ater recovery after 30
m inutes was only 15% (by weight). No m ention of oil recovery was made. The
exact frequency and power levels at which the above experiments were
4
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
conducted was not specified, although a frequency range of 2000 to 3000 MHz is
suggested. Power levels are implied to be between 300 W and 700 W.
Two publications by Fang et al. (1988, 1989), present a more detailed
analysis of batch demulsification of o il-in -w ater and w a ter-in -o il emulsions.
In their 1988 paper, emulsions of w a ter-in -cru d e oil and o il-in -w a te r in the
presence of solids were tested. In addition, field tests were conducted on waste
oil sludges. Their work with w a ter-in -cru d e oil emulsions indicated th a t, in
general, emulsions treated by microwave radiation were heated faster th an those
heated by a hotplate. Further, emulsions heated to a given tem perature by
microwave radiation often were demulsified to a greater extent than those
heated to the same tem perature by a hotplate. They concluded th at the rate of
heating is an im portant demulsification factor, and th a t the zeta potential of
the dispersed phase is also im portant. Discrepancies in their results were,
however, noted. Tests with a viscous oil—in—water emulsion with solids
indicated th at although as much as two thirds of the w ater could be recovered,
less than 50% of the oil was recoverable. Field tests on large tanks containing a
w a ter-in -o il emulsion gave favorable results. The microwave system used for
the field tests was compact and easy to use as compared to other
demulsification systems. The entire microwave treatm ent system could be
placed on the back of a truck and moved from site to site.
In 1989, Fang and his associates published another paper which recaps
much of the m aterial of the previous paper but w ith several additions. Firstly,
a vegetable oil-w ater-diatom aceous earth emulsion was tested to compare
microwave heating to heating via a hot plate. I t was found th a t if two
emulsions were heated to the same tem perature, one by each m ethod, the
5
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
emulsion treated with microwaves gave better oil recovery. Secondly, a more
detailed investigation of the effects on zeta potential due to microwave exposure
was performed. A reduction in zeta potential was indeed found in each trial,
but the reductions did not follow any reproducible pattern. Reductions in zeta
potential ranged between 5% and 20%.
Pal and Masliyah, in their 1990 paper, investigated the relative
significance of various emulsion properties with regard to microwave
demulsification. They showed th at the most im portant factor by far is the
tem perature of the emulsion. It was noted th at increased salt concentration in
w ater phase increased the rate of tem perature rise in the emulsion. The effects
of surfactant concentration was also investigated. It was found th a t for a
nonionic surfactant, an increase in the surfactant concentration reduced the
recovery. Pal and Masliyah asserted th at it was the recovery of the dispersed
phase of an emulsion which was most im portant for evaluating the effectiveness
of a particular demulsification technique. They also proposed th a t, in term s of
cost and reliability, it would be better if chemical addition could be completely
elim inated. When the comparison between microwave demulsification and
demulsification in a constant tem perature bath was made, the microwave
treatm ent was found to give improved recovery consistently. It was this batch
study th a t gave rise to the continuous demulsification study described in this
thesis.
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
CHAPTER 3
EMULSIONS
AND
D E M U L S I F I C A T I O N
3.1 W hat are Emulsions?
Emulsions are generally produced when two immiscible liquids are
mixed. The mixing may occur due to the presence of a stirrer, a pump, or flow
in a pipeline. Thermodynamically, an emulsion is unstable. All systems tend
to adopt a configuration which minimizes the energy. For two immiscible
liquids, this means minimizing the surface area between them. Since one large
droplet has less surface area per unit volume than several smaller drops, there is
a natural tendency for the smaller droplet to coalesce (ie combine on contact to
form fewer, larger droplets). Van der W aals forces a ttra c t the droplets towards
one another, increasing the num ber of contacts. The end result of this process is
the separation of the emulsion into two homogeneous and distinct phases.
Another natural process th a t assists in destabilizing emulsions is th a t of
gravity. It is often the case th at the two immiscible liquids in an emulsion do
not have the same density. This means th a t there will be a natural tendency
for the droplets in such an emulsion either to rise to the top of the continuous
phase or to sink to the bottom . The droplets become m uch more closely packed
and can therefore coalesce more easily. Even if the droplets themselves are
stable, flocculation can occur. In a flocculation process, the droplets adhere to
one another, but do not combine to create a larger single droplet. T he resulting
group of droplets is called a floe. Often flocculation occurs during a gravity
induced rising or sinking of the droplets. If flocculation occurs w ith droplets
th a t rise to the top of the emulsion (ie. the suspended phase is less dense than
7
of the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
the continuous phase), the process is called creaming. If flocculation occurs
with droplets sinking to the bottom of the emulsion, the process is called
settling. Both of these processes produce an emulsion in which there are two
distinct regions, a region of closely packed droplets with a small quantity of
continuous phase, and a region containing almost no droplets (ie. almost pure
continuous phase). While the emulsion cannot be considered well dispersed
under these conditions, it is not truly destabilized. One cannot easily recover
the suspended phase, and a little mixing would quickly restore the emulsion to a
well dispersed state.
3.2 Factors that Stabilize Emulsions
Despite the destabilizing influences discussed in section 3.1, some
emulsions can be stable for several years with no noticeable change in their
degree of dispersion. It is clear, therefore, th at there are factors which can
stabilize an emulsion against the destabilizing influences. A review of such
factors is provided by Sanders and Masliyah (1990).
One im portant stabilizing factor is the presence of solid particles in the
emulsion. Such particles tend to accum ulate and adsorb on to the phase
interface; in other words upon the surface of the droplets. The resulting layer of
particles physically impedes the droplet’s ability to approach other droplets,
because one droplet or the other m ust force its way through the particle layer in
order for contact to occur between the droplets. This type of stabilization is
shown in Figure 1. Long-chain polymers provide a particularly effective
physical barrier to droplet contact. As one can see in Figure 2(a), the polymer
is only adsorbed onto the droplet surface at a few points on the chain, leaving
8
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
the remainder of the chain extended into the continuous phase. As well as
providing a large physical barrier to droplet approach and contact, long-chain
polymers exert significant osmotic p
effects which further ham per droplet
approach and contact. This "osmotic itpulsion" occurs when two polymer
layers interact. The interaction increases the free energy of the polymer layers,
increasing the energy barrier between the droplets. A schematic is shown in
Figure 2(b).
In addition to these physical effects, ionization within the emulsion can
produce electro-repulsive effects which provide large energy barriers to droplet
approach and contact between the droplets in the emulsion. The
electro-repulsive effect occurs when ionized species of one charge are adsorbed
onto the droplet surface preferentially. This results in an excess of one charge
at the droplet surface, and an excess of the other charge in the region
surrounding the droplet. A droplet upon which this occurs is said to have an
electric double-layer. The potential of this layer is referred to as the zeta
potential of the droplet. A schematic of the electric double—layer is shown in
Figure 3. Although the net charge of the emulsion is zero, the individual
droplets possess an excess of the same charge upon their surface. The result is
like-charge electro-repulsive forces between the droplets. The energy barrier
this produces must be overcome before droplets in th e emulsion can approach
and contact one another. The strength of this energy barrier is a function of the
surface potential of the droplet , the radius of the, droplet and the perm ittivity
of the continuous phase. Another stabilizing factor for some emulsions is high
continuous phase viscosity. A highly viscous continuous phase increases the
energy required for droplet movement, and hence the energy required for
9
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
droplets to approach one another and coalesce.
3.3 Emulsion Breaking Techniques
Emulsion breaking techniques are designed to reduce the effectiveness of
one or more of the stabilizing factors discussed in Section 3.2. The breaking of a
particular emulsion usually involves a combination of techniques which is
strongly dependant on the emulsion’s characteristics.
One of the simplest and m ost effective demulsification treatm ents is to
increase the emulsion tem perature. Elevated tem peratures reduce emulsion
stability by reducing the bulk viscosity of the emulsion, the interfacial shear
viscosity in the emulsion, and decree of adsorption of stabilizing m aterial upon
droplet surfaces in the emulsion. In addition, emulsion stability is known to
decrease rapidly as the phase inversion tem perature is approached. W hen an
emulsion undergoes phase inversion, the dispersed phase and the continuous
phase become reversed; th a t is, the liquid th at originally was the dispersed
phase becomes the continuous phase of the emulsion, while the original
continuous phase becomes the dispersed phase. There are, however, a few
drawbacks to raising the emulsion tem perature. Firstly, raising the
tem perature of an emulsion can involve a significant expenditure of energy.
This is a serious problem if energy costs are high. Secondly, raising the
emulsion tem perature may cause decomposition of the emulsion components in
some emulsions. Thirdly, while raising the emulsion tem perature does reduce
the adsorption of surfactants and other particles onto droplet surfaces, it does
not directly combat stabilizing factors such as electro-repulsion.
A nother commonly employed demulsification technique is centrifugation.
10
o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
A centrifuge enhances the natural process of gravity separation, increasing the
strength of the gravitational forces acting on emulsion components of different
density
solids
This can be particularly effective when used on emulsions stabilized by
There must be a significant density difference between the emulsion
components for centrifugation to be useful. This process, however, still has no
direct effect upon electro-repulsion.
A third and far more complex method for breaking emulsions involves
the addition of various chemicals to the emulsion. Chemical addition can be
used to combat any of the stabilizing factors. There are a tremendous number
of chemicals that one can employ in breaking emulsions: chemicals to drive
particles from the droplet surfaces, chemicals to neutralize the surface charges,
etc.. Choosing the correct chemical combination is a difficult task, and requires
experienced operators. Poorly designed chemical addition can result in an
emulsion th a t is more stable than the emulsion the operator was initially trying
to break up. In addition, added chemicals m ust often be removed from the
product stream later.
11
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission .
CH APTER 4
MICROWAVES
EFFECTS
ON
AND
THEIR
MATERIALS
4.1 W hat are Microwaves?
To understand the utility of microwaves in demulsiGcation processes, one
must first understand exactly what microwaves are and can do. An excellent
review of microwave heating theory is provided in Metaxas and Meredith
(1983), and much of this chapter is from that work. Microwaves are part of the
electrom agnetic spectrum of waves, which includes radio waves, infrared
radiation, ultraviolet radiation, light, etc.. All of these waves travel at light
Q
speed (3x10 m /s in a vacuum). Electromagnetic waves exert both an electrical
and a magnetic field as they travel. Electromagnetic waves with a wavelength
between 1.0 mm and 1.0 m are referred to as microwaves. This corresponds to a
frequency range of between 300 GHz and 300 MHz. T he standard microwave
oven employs microwaves with a frequency of 2.45 GHz, giving a wavelength of
12 cm. Microwaves for commercial purposes are generally generated by a
m agnetron or a klystron. The m agnetron employs a rotating electrical field to
induce a microwave field by resonance. They are the most commonly used
commercial microwave generators. The klystron is less common. The klystron
uses a beam of electrons to amplify a weak microwave signal, and can be used in
series to give very high power output. It also gives far tighter frequency control
th an magnetrons. The prim ary disadvantage of the klystron is its cost.
Microwaves are usually transm itted using m etal tubes called waveguides.
These tubes may have a round or rectangular cross section, and are sized so
12
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
th a t the desired microwave frequency can propagate through them. Another
type of waveguide is the coaxial waveguide. Such a guide has a conducting cord
or rod down the central axis, and can transm it m; :rowave energy in a
waveguide which would otherwise be too small
4.2 Commercial Use of Microwaves
Microwaves were originally employed in m ilitary radar, but have found
many commercial applications. During radar development it was noted th at
m aterials containing w ater would heat up when exposed to microwave
radiation. This has produced numerous industrial applications in the heating
and drying fields. The discovery also initiated a significant am ount of research
into the reason for microwave heating.
4.3 Microwave Heating
The phenomenon of microwave heating is a result of the electric field
produced by microwaves. Electrically active molecules and atoms in the
m aterial to be heated will always attem pt to align themselves w ith any electric
field they are exposed to. The field produced by an electrom agnetic wave,
however, is constantly changing, completely reversing direction every
half—cycle. For a 2.45 GHz field, this means 4.9 billion times per second. The
result is a lot of movement for the electrically active molecules and atoms.
T he extent of movement depends on the molecular or atomic composition and
structure, and frequency of the electrom agnetic wave. For charged ions, simple
back and forth movement occurs. This is shown schematically in Figure 4.
B oth the movement and the resulting collisions dissipate therm al energy.
13
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission .
Another type of movement is that caused by dipolar molecules. Dipoles are
created when the electrical charges on a molecule are not evenly distributed.
The result is a molecule in which one portion is positively charged and another
portion is negatively charged. Dipoles may be either perm anent or induced.
Perm anent dipoles result from molecular geometry, and exist regardless of their
environm ent. W ater has a strong perm anent dipole created by the one-sided
arrangem ent of the hydrogen molecules. Induced dipoles result when a molecule
w ith "loose" electrons or atoms is exposed to an electric field. Helium
molecules can produce such a dipole. The field causes th e loose charges to shift,
resulting in a electrically unbalanced molecule. Dipolar molecules respond to an
electric field by rotating so th a t their positive and negative ends line up w ith
the field. In an alternating field, this causes the molecule to rotate back and
forth, em anating therm al energy and dissipating microwave energy.
A
schematic diagram of this action is shown in Figure 5. The amount of therm al
energy em anating is found to be:
P o , = 0.556x10""10 f
E*
V
av
eu rm s
(4.1)
v '
where P ay is the average power absorbed by the m aterial in w atts, f is the
frequency of the wave in hertz,
is the effective dissipation factor, Efms is
the root mean squared electric field strength in volts per m eter, and V is the
volume of m aterial in meters cubed. The wave frequency and electric field
strength are functions of the microwave generator. T he effective dissipation
factor and m aterial volume are functions of the m aterial used. The effective
dissipation factor, however, changes depending on the frequency of the
14
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
microwave and the tem perature of the m aterial. Figure 6 shows the effective
dissipation factor for distilled w ater at different tem peratures and frequencies,
Figure 7 shows d ata for 0.1 M saline at different tem peratures and frequencies.
It is clear from Figures 6 and 7 th at the effective dissipation factor is not only a
strong function of tem perature and frequency, but also is a function of water
salinity. The effective dissipation factor can be thought of as a representation
of the efficiency of a m aterial in absorbing energy from microwaves.
Another factor th a t m ust be considered when working w ith microwaves
is the penetration depth. This again depends on both th e wave frequency and
the m aterial which is being exposed to the wave. As a wave passes through a
m aterial, it is attenuated by the m aterial, reducing the effective power available
to the m aterial’s interior regions. This means th at if a m aterial is too thick, the
interior regions of th a t m aterial will not be exposed to an effective level of
microwave radiation. Some benchmark is necessary to establish the relative
ease by which microwaves of a certain frequency enter a m aterial; therefore
penetration depth is defined as the depth (ie distance from the surface of the
m aterial) at which the fraction of microwave energy available is 1/e
(approxim ately 37%), as compared to the available energy a t the m aterial
surface. This distance is usually expressed in mm. Figure 8 shows some typical
values for distilled w ater at various frequencies (Ohlsson et al., 1974).
M aterials w ith high effective dissipation factors have low penetration depths
and vice versa.
The use of microwave heating has in recent years become more
widespread due to the introduction of domestic microwave ovens. In these
ovens, a magnetron generates the microwaves. The waves are then directed to
15
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
the oven cavity using waveguides. The oven cavity is a m etal-w alled box.
Because metals reflect microwaves (although a significant current is created in
the m etal), waves entering the cavity are reflected by the cavity walls until
they strike the "load" (ie whatever foodstuff is being heated). The m ultimodal
nature of these ovens means th at they can accept and heat a wide variety of
loads. The ovens operate at 2.45 GHz with an approxim ately 2% margin of
error in the frequency. For experim ental work, the difficulty with domestic
microwave ovens is the poorly defined field configuration within the heating
cavity. This has been reported by Risman et al. (1987) and W ickersheim et al.
(1990). W ickersheim et al. showed th a t the highest field strength in a
microwave oven could be 2 to 3 times the lowest field strength, and th a t field
strengths within a microwave oven did not follow any simple pattern.
4.4 Microwaves and Emulsions
Special conditions apply when considering the microwave heating of an
emulsions, because of its heterogeneous nature. Firstly, an additional heating
mechanism is introduced. A potential difference can develop over the phase
boundaries in an emulsion. This potential difference can be acted upon by the
electric field produced by the microwaves, stressing th e boundary and
dissipating microwave energy. Secondly, the configuration of droplets of one
m aterial w ithin a continuous phase of another m aterial requires a more complex
form ula for the effective dissipation factor. One available formula is due to
Wagner (Thom as et al. 1990). W agner presented an equation for the
calculation of the complex dielectric constant c, based on the assum ption that
th e dispersed phase droplets do not interact w ith one another.
16
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
For pure substances, £ is defined as
£ = e’ - i e "
(4 .2 )
Where e" is the dissipation factor, and £’ is the dielectric constant. The
dielectric constant is also a function of tem perature and frequency, as shown in
Figures 9 and 10. One can see from these figures th a t for w ater and 0.1 M
saline, the dielectric constant is strongly dependant on tem perature and only
weakly dependant on frequency. Also one can note th a t in general the dielectric
constant is only weakly dependant on saline concentration. By dividing the
above equation by the free space dielectric constant, we can obtain the effective
values by:
W agner’s equation for the complex dielectric constant of an emulsion can
be w ritten as (Thomas et al. 1990):
(4.4)
W here
is the complex dielectric constant of the suspended phase, Cj is the
complex dielectric constant of the continuous phase, <)>is the volume fraction of
the suspended phase, and £m is the complex dielectric constant of the emulsion.
17
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
By grouping all the imagina-y terms in W agner’s equation, we can arrive at the
effective dissipation factor, c^f. This can be accomplished by employing the
standard rules for addition, subtraction, m ultiplication, and division, as applied
to complex numbers. A good reference for this information is Spiegel (1972).
A more complex direct-solution method was proposed by Rayleigh in
1892 and improved upon by Meredith and Tobias (1960). This model was
developed by considering the interaction of a cubic array of monodispersed
spheres. Using the M eredith and Tobias equation, the calculation for the
complex dielectric constant becomes:
101
2cl + £2
— 2<|) -+- 0.409
*"el + Sc2
el “ £2
2cl +£2
2 c1 + £2
+
<|> + 0.409
. £l “ £2
4£i+ 3 e2
-e-
2el + c2
£1 £2
2.133
4cl + 3f2
7
10
£1” £2
0.906
4«j +3«2
**
This equation can be used to determine the effective dissipation factor for an
emulsion by again grouping the imaginary term s. The equation has one
weakness in th at it diverges as the close packing concentration is approached
(<t> 2 0.66).
4.5 Safety w ith Microwaves
Microwaves have significant danger associated with their use ( see
M etaxas and M eredith 1983). The prim ary danger from these invisible waves is
th e adverse heating of body parts by inadvertent exposure to them. There are
also suggestions th a t there may be long term genetic effects from exposure. For
a batch system, such as a domestic oven, it is quite simple to prevent any
18
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
microwave exposure. The only source of leakage during normal operation is at
the door seal. In most ovens this problem is solved by only allowing the oven to
operate when the door is fully closed, and by installing quarter wavelength
chokes around the door. Q uarter wavelength chokes are L -shaped grooves
which trap microwaves by redirecting them towards the "dead end" of the
groove. A conventional microwave oven cannot be sold if the microwave
o
leakage exceeds 1.0 m W /cm , one ten th of the North American maximum long
term exposure limit. For a continuous system, leakage prevention is
complicated by the need for inlet and outlet ports. If these ports can be made
small enough, a cutoff waveguide can be used. This is a waveguide tube th a t is
too small to permit the microwave passage. Provided th a t the cutoff waveguide
is of a minimum length (which depends on port size and wave length), leakage
can be almost eliminated. If the ports m ust be larger, quarter wavelength
chokes, absorptive port wall m aterials, and/or corrugated chokes must be used.
19
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
CH APTER 5
DEVELOPM ENT OF A THERM AL MODEL
The analysis of the tem perature profile of a flowing liquid in a coil
subjected to microwave heating can be achieved by using a mass and energy
balance on a differential volume element of the flowing liquid. A diagram of
the elem ent used is found in Figure 11. It is assumed th a t the coil tube radius
is very much smaller than the coil length, and th a t losses from the coil surface
are negligible. As a result no radial tem perature variation is assumed. This
leads to a one-<iimensional model. The stea d y -sta te heat balance on the liquid
element is given by:
R ate of energy output by the fluid - R ate of energy input by the fluid =
R ate of microwave energy absorption - R ate of energy loss to cavity (5.1)
The model for the axial tem perature variation of a flowing emulsion is
considered below for two cases
The first case is for the non-boiling condition
where the emulsion does not reach its boiling point. T he second case considered
was th a t of the boiling condition, where the emulsion reaches its boiling point.
In both cases it is assumed th a t the conduction in the liquid phase and along
the coil walls is negligible.
5.1 T he Non—Boiling Condition
If we apply the energy balance to non-boiling conditions, the term s may
be given a m athem atical representation. The rate of energy input by the liquid
is:
20
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
(5.2)
H = rti C p T
where rh is the mass flowrate in kg/s, Cp is the liquid heat capacity in
J /k g °C, and T is the liquid tem perature in degrees Celsius (°C). The energy
output by the liquid is given by:
H + dH = rti C p T + ih Cp dT
(5.3)
Conservation of mass insures th at drti = 0 . We can reduce the left side of the
equal sign in equation (5.1) to :
m Cp dT
(5.4)
M etaxas and M eredith (1983) derived the power dissipated w ithin a
m aterial subjected to microwave radiation (see equation 4.1). The microwave
power absorbed by th e liquid in the element is given by:
0.556* 10 10 f x R2 E2m se2ff dx
(5.5)
The constant 0.556* 10- 1 ® in the above equation, the result of m ultiplying the
free space dissipation factor by 2x, acts as a unit conversion. T he quantity f is
the microwave frequency, expressed in Hz. The element ra d iu s , R, is expressed
in meters. The term E rms stands for the (root mean squared) electrical field
strength and is expressed in V /m . The quantity
is th e effective dissipation
factor. It is a representation of the ability of a m aterial to absorb energy from
21
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
microwaves. Finally, x is the axial distance from the start of the coil (where
microwave exposure begins).
The rate of energy loss to the cavity can be expressed as:
2 * R U ( T - T o) d x
W here
(5.6)
is the cavity tem perature in °C and U is the overall heat transfer
coefficient in W /m
2o
C. Combining the above equations leads to the overall
energy balance equation:
* Cp d T /d x = 0.556* 10"10 f * R2 E2ms ejff - 2 * R U (T - T J
(5.7)
One can simplify equation (5.7) by using microwave exposure time. The
exposure tim e is given by:
o
pT X R
x
r = _ ± _____________
lb
(5.8)
Q
where p^ is the liquid density in kg/m
. Making use of equation (5.8), the
energy balance equation becomes:
A t C p « / d r = 0.556*10-1 0 f f?tm a <”f( _ _ y L
(T - T J
(5.9)
This is the energy balance equation for the fluid element under non-boiling
conditions. Note th a t conduction along the coil, either in the fluid or in the coil
22
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
tube walls, has not been included in this formulation.
Equation (5.9) was derived for a single fluid. For the case of an
emulsion, the values of p ^, Cp , and
must be calculated differently, p^ is
evaluated by calculating a volume average between the two emulsion
components. The expression for p^ becomes:
Pi =
(5,1°)
3
where p^ is the density of the dispersed phase in kg/m , p2 is the density of the
O
continuous phase in kg/m , and <J>is the volume fraction of the dispersed phase.
To obtain C , a weighted average of the pure component values is used. The
I*
expression for Cp becomes:
CP=cpi» +V 1-*)
(511>
where Cp l is the heat capacity of the dispersed phase in J /k g °C , Cp2 is the
heat capacity of the continuous phase in J /k g °C , and w is the weight fraction
of th e dispersed phase,
is calculated using equation (4.5).
As noted in C hapter 4, the electric field strength is poorly defined w ithin
a domestic microwave oven. It is therefore necessary to estim ate the strength of
th e field in some fashion. It was therefore assumed th a t the electric field was
constant along the length of the coil. The actual strength of the field was
estim ated by a trial and error procedure based upon the coil inlet and coil outlet
tem perature of the emulsion.
23
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
5.2 The Boiling Condition:
Under boiling conditions, additional equations are needed to describe the
system. Once boiling tem perature is reached, the tem perature of the liquid
remains constant, and the volume of the liquid in a coil element changes. This
is im portant because liquid water absorbs microwave energy far more effectively
th an steam (to the point th a t steam can be considered non—absorbing).
F urther, in a continuous flow system, the flow changes from single phase flow
to tw o-phase flow.
To modify the heat balance for use under boiling conditions, two changes
m ust be made. Firstly, the constant liquid tem perature means th a t the rate of
energy gain by the element cannot be expressed by equation (5.2). which
represents simple heating of liquid water. Instead, the energy input of the fluid
can be expressed as:
(5.12)
where H is the enthalpy of the phase at boiling. The subscripts refer to the
liquid phase, L, and the vapor phase, v, respectively. The energy output of the
tw o-phase fluid can be given by:
(5.13)
HL a L + d( h l * L ) + Hv my + d( Hy my )
By considering the mass balance we can see th at dmy = -d m ^ . W e can
consider
and Hy to be constant. If we let A^y = Hy -
, we can reduce
th e left side of the equation to:
24
R ep ro d u ced with p erm ission o f th e copyright ow ner. Further reproduction prohibited w ithout p erm ission .
where A^v is the specific heat of vaporization in J/k g . The second modification
th a t is required concerns the rate of energy absorption by the fluid in the
element. When water is under boiling conditions in a given element, both water
and steam occupy the element. Consequently, the absorption term must be
modified to give:
0.556*10 10 f E j m s <effa d r
(5-15)
where a represents the fraction of the element occupied by the liquid. Chisholm
(1986) gives an expression for a as:
"(jr/7;')+Rn-xr7*T
—
where x >s the mass dryness fraction and K is the phase velocity ratio, p ^ and
py represent the densities of the liquid
liqu and vapor phases, respectively. The
mass dryness fraction is defined by:
X= —
rh
(5.17)
C alculating the phase velocity ratio is more difficult. It has been shown th a t
for a wide range of conditions, K may be evaluated from:
25
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
if X > 1.0, then K =
(5.18)
if X < 1.0, then K =
(5.19)
The Lockhart-M artinelli param eter, X, is given by:
I
X =
Where
0 —x
(5.20)
is the vapor density in kg/m . V e also need to know the
homogeneous density of the m ixture (th at is, the density when K =1.0). The
homogeneous density is given by:
1 .0
(5.21)
^h =
This allows us to calculate a as it is needed in equation (5.15). The final energy
balance equation for boiling conditions can be w ritten as:
dih
,v -r
I
= 0 .5 5 6 M 0 '10 f E^mfi <”ff a _
J
”
(T - T J
(5.22)
R
There is a further special consideration when attempting to model an
emulsion. Since the two liquids making up the emulsion boil at different
26
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
tem peratures, one liquid will boil before the other. This means th at under these
conditions only one liquid will be boiling. For oil and w ater this is usually the
water. The vapor pressure of the oil is considered to be negligible, therefore
boiling can be presumed to occur at the boiling point of the water. Hence, a
may be considered to apply only to the water phase.
The mass of vapor being generated is not known, however. As this
information is required in order to calculate E rmg under boiling conditions,
equation (5.22) cannot be used to predict the internal tem perature profile.
Instead, the model becomes a means of curve fitting the experim ental data. By
assuming a value for Erm s, a tem perature profile can be generated. This profile
can then be compared to the experim ental d ata
By altering the value of Ermg
used, one will eventually generate a tem perature profile which m atches the
profile of the experim ental data. We may then determine the am ount of vapor
generated from the model.
We now have an energy balance equation for each region in the coil.
Equation (5.9) represents the energy balance under non-boiling conditions and
equation (5.22) applies under boiling conditions. The switch from one equation
to the other occurs at the boiling tem perature of the liquid. The numerical
procedure used was a combination of the Runge—K u tta stepping technique w ith
a simple shooting formula (Gerald and W heatley,1985). T hat is, the profile was
to be developed by using some initial estim ate of the root mean squared electric
field strength. A R u n g e-K u tta formulation of equation (5.9) is used until the
boiling point is reached. Then the R u n g e-K u tta formulation is again employed,
this tim e on equation (5.22). Equation (5.22) completes th e m arch along the
coil (if necessary). The final output is then compared to the observed outlet
27
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission .
conditions and if it is not close enough, a new estim ate of G
is made and
° ’
rms
process is repeated.
28
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
CH A PTER 6
MODEL VALIDATION THROUGH COM PARISON W ITH
EXPERIM ENTAL DATA
6.1 EXPERIM ENTAL APPARATUS
A household microwave, operating a t 2450 MHz and a m axim um rated
power output of 720 W, was employed. To allow the te st fluids to flow through
the oven cavity, a glass coil 15.3 m long w ith a nominal internal tube diam eter
of 5 mm was constructed. The internal diam eter of th e coil was 0.12 m . A
straight section of glass tubing was attached to both the inlet and outlet of the
coil. Two holes were drilled in the rear facing o f the oven cavity. Each hole
was 13 mm in diam eter and was affixed with a section of copper tubing 29 m m
long and of identical diam eter (i.e. 13 m m), extending from th e outside oven
wall away from the oven cavity. This tubing is referred to as a cutoff
waveguide and greatly restricts the microwave leakage from th e hole. The glass
coil cou’d now be positioned in the microwave cavity, w ith th e straight inlet
and outlet sections extending through th e choked holes to the microwave oven
exterior.
To allow tem perature measurements along the length of the coil, three
rubber stoppered ports were built into the glass coil. T hree holes 2 mm in
diam eter were drilled in the cavity top directly above th e port positions on the
coil. No protective measures (i.e. chokes or cutoff waveguides) were required as
the holes were very small. Type J thermocouples were inserted through th e
cavity top and then through the rubber stoppers. Two additional
thermocouples were placed in the system. One therm ocouple was placed to
29
p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
measure the inlet tem perature of the test fluid and the second was placed to
measure the outlet tem perature. Placem ent of these two thermocouples was
particularly im portant. The thermocouples are placed outside of the inlet and
outlet cutoff waveguides. Placem ent within the waveguides produces a coaxial
waveguide which will transm it significant amounts of microwave energy. All of
the thermocouples were grounded to prevent arcing due to microwave exposure.
M easurements from the thermocouples were taken by a six channel
thermocouple panel m eter
A C ole-Palm er M asterflex peristaltic pump was used to pump the test
fluids through the coil. Fluids were pumped from a 2 litre graduated cylinder
into the coil. At the outlet, a three-w ay valve perm itted the outgoing fluid to
be directed to either a 500 ml graduated cylinder or another 2 litre graduated
cylinder.
To reduce heat losses from the coil during heating, the coil was wrapped
in fiberglass insulation. A schematic of the complete system is provided in
Figure 12. A Bach—Simpson microwave leakage detector was used to measure
microwave 1
ge and ensure lab safety.
6.2 M ICROW AVE HEATING O F W A TER AND SALINE
To test the developed theoretical model, a series of experim ents was
performed w ith distilled w ater and w ater containing known quantities of NaCl.
Experim ents covered salt concentrations from 0.0 M to 0.12 M. The flowrate of
the solutions in the coil ranged from 0.000648 L/s to 0.0029 L /s. All the
experim ental runs were performed a t the highest power setting of the
microwave oven, to avoid any tem perature cycling as observed by K urda et al.
30
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
(1989).
For this series of experiments, the physical properties of water were
taken as constant and independent of tem perature. Therefore Cp was taken as
4180 J /k g °C , p^ as 1000 kg/m**, and
as 2.257x10® J/k g . The effective
dissipation factor was calculated for each point on the coil by interpolating data
from the functions shown graphically in Figures 6 and 7. The ovens operating
frequency was 2.45 GHz. Because of the difficulties involved in defining the
electric field strength in a commercial oven, the field strength was taken as
constant throughout the oven, and was evaluated based on the inlet and outlet
conditions of the fluid.
To evaluate heat losses from the coil, experiments were conducted under
batch conditions where the coil was filled with distilled w ater and then heated
to a high (but not boiling) tem perature by the microwave oven. The w ater in
th e coil was then allowed to cool and the cooling rate was recorded. Figure 13
shows representative cooling curves. The overall heat transfer coefficient, U,
was then calculated from these experiments and it was found to be
0.0062 W /m 2 °C. This corresponds to a rate of heat loss of about 2 W for the
entire coil. W hen compared to overall heat gains by th e liquid of 400 to 500 W,
this is a negligibly small number. As a result, it was decided th a t
would be
taken as a constant (25°C) when employing the model to predict tem perature
profiles. This simplifies the calculation and has little influence on the results.
The first series of experiments was performed w ith the condition th a t the
outlet tem perature from th e coil would not reach boiling tem peratures. In
Figures 14 through 19, comparisons between d ata recorded in the experiments
and th a t predicted by the model, based on inlet and outlet tem peratures, is
31
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission .
presented. The axial locations in the coil are converted to exposure times using
equation (5.8), since exposure tim e is directly proportional to axial distance for
a given coil radius, liquid mass flow rate, and liquid density. Note th at because
the effective dissipation factor is a function of tem perature, the various runs
could not be collapsed onto one curve.
In Figures 14 through 17, the tem perature variation with exposure time
at a liquid flow rate of about 0.0016 L/s for different NaCl concentrations is
shown. In all cases, the model prediction is fairly close to the experimental
data. An exact m atch is not expected as the eiactric field strength within the
cavity is not known and was assumed to have a constant value. The total
power absorbed by the fluid flowing through the coil ranged from 448 W for
distilled w ater to 509 W for a 0.12 M saline solution. In Figure 18, the
experimental variation of tem perature with exposure tim e, together with the
model prediction at two different saline concentrations are shown. One can
observe th a t both the d ata generated by experim entation and the model
prediction are grouped closely together. In Figure 19, the experim ental data
and the model prediction for a very high flowrate are shown. Again, a close
agreement between the experim ental results and the model is observed.
A second set of experiments was performed where boiling tem peratures
were reached by th e fluid. Figures 20 through 23 show the variation of saline
solution tem peratures w ith exposure time for various saline concentrations
under these conditions. A best fit tem perature profile using the model equation
is also shown in each of these figures. In each case, we observe th a t th e fluid
tem perature rises until the boiling tem perature is reached, after which the fluid
tem perature remains constant. The fit between th e model and the d a ta is very
32
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
good in all cases. In Figure 23, the model profile and the experimental d a ta for
the two fluids at the same flowrate but different saline concentrations are
shown. In this case we note th a t the tem perature profile for the distilled w ater
run is very different than th at for 0.06 M saline. W hen one compares the strong
saline dependence of the tem perature profile in the boiling case to the very weak
saline dependence of the tem perature profile in the non-boiling case
(Figure 18), the fundam entally different nature of boiling vs non-boiling
systems is apparent. For distilled w ater, power absorption a t boiling
tem peratures is less than 20% of the values at coil inlet tem peratures. The loss
of liquid due to vaporization further reduces the am ount of power absorped per
coil element in this region. The to tal power introduced into the oven, however,
does not change. Power not absorped in the boiling region of the coil m ust be
either absorped by the oven walls, or, in more absorptive regions of the coil. As
a result the tem perature in th e early portion of the coil rises at a more rapid
rate under boiling conditions than under non—boiling conditions. For th e saline
solution, the loss oi liquid due to vaporization also reduces the boiling regions
ability to absorp power. However, as w ater is vaporized, the liquid saline
concentration becomes higher, increasing power absorption ability of the
remaining liquid. It appears th a t these two processes m aintain a nearly
constant power absorption ra te in all portions of the coil for 0.06 M saline. This
absorption profile is similar to th a t of the non-boiling 0.06 M saline, hence the
sim ilar boiling and non—boiling tem perature profiles for this liquid.
In Figures 24 and 25, the power absorption variation w ith flowrate for
the cases of 0.06 M saline and distilled w ater , respectively, are shown. The
d a ta are for boiling and non—boiling cases. For th e case of non—boiling, the
33
o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
power absorption is simply calculated from the difference between the inlet and
outlet solution tem perature. For the case of boiling, the power absorption is
th a t given by the model when a best fit to the experimental tem perature d ata is
achieved. The figures clearly show th at for experimental runs conducted under
boiling conditions, the power absorption increases as flowrate increases. Under
non-boiling conditions, a relatively constant power absorption, regardless of
flowrate, is observed. The difference between the two cases can be directly
related to the relative volumes of water in the coil. The ability of w ater to
couple (i.e. absorb) energy from the microwaves in the oven before the energy is
dissipated into the oven walls is strongly dependant on the am ount of liquid
w ater present in the oven. Since the boiling process converts w ater from its
liquid (i.e. microwave absorbing) phase, to a vapor (i.e. nearly microwave
transparent) phase, the amount of w ater present under boiling conditions will
be less the longer the water is heated. Consequently, the am ount of energy
absorbed is diminished for boiling systems (D atta 1990). This explains why
under boiling conditions the total power absorbed decreases as th e flowrate
decreases. As more and more w ater is vaporized, the ability of the remaining
w ater to couple energy is diminished. This coupling effect also explains why the
to ta l energy absorbed by the non-boiling distilled w ater is less th an th a t
absorbed by non—boiling 0.06 M saline solution. As can be seen on Figure 26,
th e presence of NaCl increases the w ater solution ability to couple energy from
microwaves, i.e. saline solutions can couple more energy th an distilled w ater.
These effects apply for the liquid volumes used in this experiment. At much
larger volumes, the total energy absorbed reaches a lim iting value (based on
microwave design).
34
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
CH APTER 7
EX PERIM ENTS IN CONTINUOUS M ICROW AVE DEM ULSIFICATION
7.1 EXPERIM ENTAL SETUP
The experiments in continuous microwave demulsification where
conducted using the apparatus described in Section 6.1 of C hapter 6. In each
group of experiments an emulsion type was tested. The param eters investigated
were those of dispersed phase recovery and tem perature profile within the coil.
The definition of recovery is needed for both the saline solution phase
and the oil phase. Recovery of the saline solution is defined as:
Volume of saline reco v ered
(Volume fraction of s a l i n e in original em u lsio n )(T o tal sample volume)
Similarly, the recovery of the oil phase is defined as:
Volume of o i l R eco v ered
(Volume fraction of o i l i n o rig in al em u lsio n )(T o tal sample volume)
Recovery was measured by taking a 300 ml to 400 ml sample of th e emulsion
leaving the coil and allowing it to settle for 30 min.
Two different oils were used in the emulsions tested: Bayol-35 and
bitum en. Bayol-35 is a very light mineral oil having a density of
approxim ately 850 kg/m
and a viscosity about 2.4 tim es th a t of w ater at room
tem perature. Bitumen is a very heavy crude oil having a density of
35
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
2
approxim ately 1070 kg/m
and a viscosity of 3500 tim es th at of water at room
tem perature. Its viscosity at 55°C is 10 times th at of w ater and at 100°C it is
0.5 tim es th a t of w ater (Singh et al. 1989).
Two different surfactants were used in the emulsions tested:
Em sorb—2500 and T riton X-100. Em sorb-2500 is a non-ionic oil soluble
surfactant, and was used to make the w ater-in -o il emulsions. T riton X -100 is
a non-ionic water soluble surfactant, and was used to make all o il-in -w a te r
emulsions
Droplet diam eter was measured for the saline—in—Bayol emulsions and
the b itu m en -in -salin e emulsions and was found to average about 30 jun in both
cases. O ther tests showed that mixing tim e had no effect upon the final
recovery w ithin the range of 5 to 15 minutes (which was the range of mixing
tim es used).
7.2 DEM ULSIFICATION OF SALINE—IN -B A Y O L EMULSIONS
A set of experim ental runs were conducted using a salin e-in -o il
emulsion. T he oil used was Bayol-35. Emulsions w ith varying dispersed phase
concentrations were prepared as follows. A known am ount of Bayol—35 was
added to a large beaker and stirred using a G ifford-W ood Homogenizer.
Sufficient surfactant was added to give the final emulsion a surfactant
concentration of 0.1%. The surfactant used was Em sorb-2500. A 0.06 M saline
solution was then added in sufficient quantity to give the required concentration
of dispersed phase ( the 0.06 M saline solution ). Concentrations are reported
as percent of total volume. The emulsion was m ixed by the homogenizer during
all additions and for a t least 5 minutes after all additions had been completed.
36
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission .
To interpret the results of a test, one must be familiar with the
characteristics of the resulting emulsion. One notes th a t the saline solution is
the dispersed phase while Bayol-35 is the continuous phase. The density of the
saline solution may be taken as 1000 kg/m , while th a t of a light mineral oil
like bayol is approximately 850 kg/m . This means th a t the saline droplets will
naturally settle, leaving a clear oil layer on top of th e emulsion. Since we are
interested in the coalescence of the saline water droplets ( which represents true
demulsification), we cannot use the appearance of a clear oil layer as an
indication of the demulsification of the emulsion. R ather, we m ust look for the
development of a clear saline layer at the bottom of the emulsion. This layer
can only be produced by coalescing saline droplets and hence demulsification.
The various stages of a w ater—in-B ayol emulsion are shown in Figure 27.
Demulsification tests on the emulsions were conducted over a range of
flowrates and with volumetric saline concentrations between 10% and 50%.
Figure 28 shows the results plotted against the tem perature of the emulsion at
the coil outlet. One can observe th at the recovery of saline increases w ith
increasing outlet tem perature. Similarly, Figure 29 shows th a t when recovery is
plotted against the initial tem perature in the m easuring vessel, where the
emulsion is collected, recovery of saline increases w ith the tem perature of the
measuring vessel. Figure 30 shows the variation of recovery w ith flowrate. One
can observe th a t the recovery of saline increases vr*h decreasing flowrate. The
highest saline recovery achieved for this system was less than 0.5.
The model developed in C hapter 5 was tested for the evaluation of the
tem perature profile of th e flowing emulsion. The density of the oil was taken as
850 kg/s, and its heat capacity, Cp, was taken as 1470 J /h g °C . T he effective
37
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
dissipation factor, e’’^, for Bayol-35 was taken to be 0.002, and the dielectric
constant, e\ was taken to be 2.201. The latter value was found by
experim entation using experimental equipment in the departm ent of Electrical
engineering developed by Xi (1992). The effective dissipation factor was
estim ated by noting the values for similar oils listed in Von Hippel (1954). The
physical and dielectric properties for saline solutions were the same as those
employed in the pure w ater and saline experim ents of section 6.2 in C hapter 6.
The dielectric properties of the oil were very low relative to those of saline and
were therefore taken as tem perature independent. The M eredith and Tobias
equation (4.5) for calculation of emulsion effective dissipation factor was used.
Model predictions using the outlet tem perature were compared w ith the
observed tem perature profile within the coil. One such comparison is shown in
Figure 31. As one can observe, a surprisingly close agreem ent is obtained.
7.3 DEM ULSIFICATION OF B A Y O L -IN -SA LIN E EMULSIONS
A set of experim ental runs was conducted using an o il-in -salin e
emulsion. The salt content of the saline solutions was 0, 0.06, and 0.12 M. The
emulsions were prepared as follows. Saline solution o f the required salt
concentration was added to a large beaker and stirred by a G ifford-W ood
Homogenizer. Surfactant was added so as to give the final emulsion a
surfactant concentration of 0.1% by volume. The surfactant used was Triton
X -100. Then the oil (B ayol-35) was added in a sufficient q uantity to give the
final emulsion the desired oil concentration. The emulsion was mixed by the
homogenizer during all additions and for a t least 5 minutes after all additions
had been completed.
38
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
For these emulsions, the dispersed phase was the oil and the continuous
phase was the saline solution. This means th at the dispersed phase was lighter
than the continuous phase, hence creaming can be expected to occur. The
critical recovery therefore is th at of the oil. The various stages of a
b ay o l-in -w ater emulsion are shown in Figure 32.
The first set of experiments conducted with the o il-in -w a te r emulsions
was similar to those performed for the saline—in-oil emulsions. A saline
solution of 0.06 M was used to make the emulsions. Tests were conducted at
various flowrates and with volumetric Bayol—35 concentrations between 10%
and 50%. The results are shown in Figure 33. As one can observe, there is
virtually no oil recovery for all of the outlet tem peratures.
Another set of experiments was performed w ith the Bayol—in—water
emulsion type. In these tests the volumetric Bayol-35 concentration was held
at 50% and the saline concentration was varied between 0.0 M and 0.12 M. The
results are shown in Figure 34. One can observe th a t the saline concentration
appears to have little effect upon final recovery.
The tem perature profile of the emulsion was used to test the model
developed in C hapter 5. A comparison between an experim ental run and the
model prediction is shown in Figure 35. W e again obtain close agreement
between the model predictions and the experim ental measurements.
7.4 DEM ULSIFICATION OF B ITU M EN -IN -SA LIN E EMULSIONS
A set of experim ents was conducted tor b itu m e n -in -sa lin e emulsions.
The salt concentration of the saline solution was 0.06 M. T he emulsion was
prepared as 'ollows. A beaker containing 0.06 M saline solution and a beaker
39
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
containing bitumen were heated to 55°C in a constant tem perature bath. A
known quantity of the 0.06 M saline solution was added to a large beaker and
was then stirred by a G ifford-W ood homogenizer. Surfactant was added so
th a t its concentration in the final emulsion would be approxim ately 1.0% by
volume. The surfactant used was Triton X-100. The heated bitumen was then
added in a sufficient quantity to give the final emulsion the correct volumetric
concentration of bitumen. The emulsion was mixed by the homogenizer during
all additions and for at least 5 minutes after all additions had been completed.
For this emulsion, bitum en is the dispersed phase and saline is the
continuous phase. The density of bitumen is 1070 kg/m** at 25°C, making it
denser than the saline. One can therefore expect settling of the bitum en
droplets to occur. The im portant recovery is therefore th a t of bitumen. The
various stages of a b itu m en -in -w ater emulsion Figure 36.
The experiments performed with b itu m e n -in -w ate r emulsions were
similar to those for the saline-in-B ayol emulsions. The recovery of bitum en
and saline are shown in Figure 37. The recovery of the dispersed phase for
bitum en—in -saline emulsion is different from those of the previous emulsion
types. The recovery of bitum en is tem perature invariant over the range of
outlet tem peratures tested (80°C to 105°C), with a value of about 0.35.
The bitum en emulsions were also used to test the developed model. The
Q
density of the bitum en was taken as 1070 kg/m , and the heat capacity of
bitum en was taken as 1570 J /k g °C. The dielectric properties for bitum en were
determ ined by X i’s (1992) testing equipment. The dielectric properties were
found to be e ^ = 3.437 and t “ff = 0.046. As before, physical and dielectric
properties for saline were taken as those used in Section 6.2 in C hapter 6. The
40
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
Meredith and Tobias equation (4.5) was used. An example of the tem perature
profile is shown in Figure 38. Once again, the agreement is very good between
the experimental d ata and the model prediction.
7.5
Conventional Demulsification Equipment
A large constant tem perature bath was constructed from a m etal basin
measuring 0.52 m X 0.52 m X 0.29 m. Two auxiliary heaters, with power
outputs of 0.5 kW and 1.5 kW, were placed this large constant tem perature
bath. These heaters were to increase the rate of heating w ithin the bath and to
help maintain bath tem perature when the desired tem perature was reached
within the bath. A small constant tem perature bath containing a 0.5 kW
heater, a stirrer, and a tem perature controller, was linked to the larger b ath by
tubing. W ater flow from the small bath to th e large b ath was m aintained by
the stirrer which acted as a pump, forcing water from the small b ath through a
section of tubing into the large bath. W ater flow from the large bath to the
small bath was m aintained by siphoning action through a large diam eter section
of tubing. The tem perature controller in the small bath acted by turning the
heater in th a t bath on or off, as necessary. The heat losses from the constant
tem perature baths made it necessary to use one or botu of the auxiliary heaters
in order th a t the required tem peratures be m aintained. B oth baths were well
insulated to minimize heat losses.
A coil identical to th at used in the microwave demulsification
experim ents was immersed iv. the larger constant tem perature bath. A
peristaltic pump was used to pump the emulsion. The emulsion was pumped
from a 2 litre graduated cylinder and samples were collected in a 500 mL
41
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
graduated cylinder. W aste emulsion (i.e. emulsion not taken as part of the
sample) was pumped to another 2 litre graduated cylinder. A schematic of the
experim ental se t-u p is shown in Figure 39.
7.6 Conventional Demulsification of B itum en-in-S aline Emulsions
Emulsions were prepared using the procedure described in Section 7.4.
The constant tem perature bath was brought to the desired tem perature (a
process requiring 3 to 4 hours). The b itu m en -in -salin e emulsion was then
pum ped through the coil at a known flowrate. The emulsion tem perature along
th e coil was recorded, along w ith the recovery of bitum en and saline achieved.
The recoveries are shown in Figure 40. Clearly, Figure 40 shows th a t there is a
large difference in the bitum en recovery between the conventional and
microwave heating techniques. At near boiling tem peratures, the recovery of
bitum en using conventional heating is significantly higher than th a t
accomplished w ith microwave heating, while a t lower tem peratures, the
recovery of bitum en from conventional heating is com parable to th a t of
microwave heating. From Figure 41 one can note th a t the tem perature history
of the emulsion in the two modes of heating is very different. For the
conventional heating mode, the emulsion attains its m axim um tem perature less
th a n 4 m from the inlet of the coil and m aintains this tem perature until its
outlet. The microwave heating mode, however, heats the emulsion in a near
linear fashion from inlet to outlet. Consequently, the microwave heated
emulsion only experiences high heating tem peratures in a small region near the
end of the coil. Pal and Masliyah (1991) showed th a t for batch systems, the
demulsification tem perature is critical. For Bayol—in—water emulsions, it was
42
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
found that oil recovery improved almost exponentially once the emulsion
tem perature exceeded 70 °C. A plot from Pal and Masliyah is shown in Figure
42. The information provided by Figures 40 to 42 shows th a t the microwave
system used in this experiment does not provide more effective demulsification
as compared to the conventional heating system used in the experim ent. .
7.7
Conclusions
1.
For the case of a non-boiling fluid, a simple model was developed
to predict the tem perature variation of an emulsion flowing in a
coil due to microwave heating. The model predictions and the
experimental d ata indicate th at the tem perature gradient along
the coil is nearly linear, regardless of salt content.
2.
For the case of a fluid which reaches boiling tem perature, tht:
tem perature profile is strongly dependant on the salt cont ji
3.
Power absorption for an emulsion is a function of b oth flcv/rate
and saline concentration if boiling tem perature is reached. Under
non-boiling conditions, power absorption is only a function of
saline concentration.
4.
For the case of a saline-in-oil(B ayol-35) emulsion, saline
recoveries of close to 50% were obtained.
43
o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission .
5.
For the case of an oil(Bayol—35 )-in —saline emulsion, poor oil
recoveries were obtained. Oil recovery was insensitive to saline
salt content.
6.
For the case of bitum en—in-saline emulsion, a tem perature
insensitive bitum en recovery of about 35% was obtained.
7.
For the case of b itu m en -in -salin e emulsion, the conventional
heating mode was found to give better recovery of bitum en than
the microwave heating mode for the systems used in this
experiment. The conventional heating mode reached high
tem peratures more quickly than the microwave heating mode.
Batch demulsification studies indicate th at the emulsion
tem perature plays an im portant role in the demulsification
process. A redesigned im
wave system th a t can provide a
higher tem perature over a greater portion of the coil could prove
advantageous.
44
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Figure 1:
Particle deposition on an emulsion droplet
45
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
(oO
Figure 2:
Polym er chain interactions between emulsion droplets resulting
in stabilization of the emulsion:
a}physical blockage of droplet contact
bjosm otic repulsion due to chain interactions
46
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
Figure 3:
The electric double layer on an emulsion droplet
47
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Figure 4:
Movement of charged particles in a microwave induced electric
field
48
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
+
4*
+
Figure 5-
Movement of dipolar species in a microwave induced electric
field
49
R ep ro d u ced with p erm ission of th e copyright ow ner. Further reproduction prohibited w ithout p erm ission .
(
FACTOR
DISSIPATION
EFFECTIVE
TEMPERATUP.S f C )
Figure 6:
Temperature dependence of the effective dissipation factor for
distilled water
50
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
100
SO h
£ eo
C * 4 5 0 MHz
• = 90 0 MHz
V = 2 6 0 0 MHz
C 'J
EFFECTIVE
DISSIPATION
FACTOR
(
ViJ
o
20
60
SO
100
120
TEMFZRATU5E (°C)
Figure 7:
Frequency dependence of the effective dissipation factor for a
0.1 M saline solution
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
(m m )
PENETRATION
DEPTH
100
0
•
20
- 91 5 MHz
40
60
£0
TEMPERATURE (°C)
Figure 8:
Temperature dependence of microwave penetration depth for
distilled water at various frequencies
52
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
9C
DIELECTRIC
CONSTANT
(e*
80
0
20
40
60
80
100
TEMPERATURE (°C)
Figure 9:
Tem perature dependence of the dielectric constant fcr distilled
w ater and 0.1 M saline
53
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
■ w a te r
= s a lin e
V
» w a te r
r
= s a lin e
□
w a te r
■ * s a lin e
at
at
at
at
at
at
4 5 0 MHz
45 0 MHz
900 MHz
900 MHz
2800 MHz
2800 MHz
DIELECTRIC
CONSTANT
(e ’
0
•
Figure 10:
Frequency dependence of the dielectric constant for distilled
water and 0.1 M saline
54
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
mCpT
mCpT + d(mCpT)
H Ln\. + Hrh,
HLmL+ H ,m v
^ ( H j n L) + d C H m ;
0.556*lo'° f-rrR Erms
2 Sefr
dx
Figure 11:
2 R U (T - T J dx
Elem ent for evaluation of heat and mass transfer w ithin a coil
heated by microwave irradiation
55
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Graduated
cylinder
_J
Graduated
cylinder
pump
1
i
©
&
L . „ .
i
!
D
0
K
©
>
Figure 12:
!
L
i
1
m
-
f
*
i
Schematic diagram of continuous microwave heating apparatus
56
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
50
45
V - th erm o co u p le 2 i -i
0> = th e rm o c o u p le 3
A = th e rm o c o u p le 4
I
1-
7 -v -v
w
( C)
fly
TEMPERATURE
NaCl c o n c e n tr a tio n :
0 .0 1 7 1 M
v -v
V
\
\
y o e > A-A
40
-
o o \.
v
\
a -a .
\
\
A-A
V-V
A-A
>
a -a
0 -0\
JO
VN
v -v .
A\
0 -0 .
O-O.
A-A
V\
V-V.
A -A -A
V
oo
< X H 0\
r
L
Ou
a-aV
I
r
0
20
40
60
tu g :
Figure 13:
80
A
<x>o.V
100
20
( m in u t e s )
Variation of temperature with time for an insulated coil
containing microwave heated 0.0171 M saline solution
57
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
12C
-
TEMPERATURE
( C)
100
O
= d a ta a t 0 .0 0 1 6 L /s
m o d el e s tim a te of
p r o file b a s e d o n o u t ­
le t te m p e r a tu r e
av era g e pow er a b so r b e d
440 W
120
180
240
EXPOSURE TIME (SECONDS)
Figure 14:
300
:
36 0
Experim ental and model based tem perature profiles for
distilled w ater flowing through a continuous microwave
heating system at 0.0016 L/s (m easured a t the inlet
tem perature)
58
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
TEMPERATURE
( C)
1 00
O
= d a ta a t 0 .0 0 1 6 L /s
m o d e l e s tim a te of
p r o file b a s e d o n o u t ­
le t te m p e r a tu r e
average pow er ab sorb ed :
484 W
60
Figure 15:
120
180
240
EXPOSURE TIME (SECONDS)
300
360
Experim ental and model based tem perature profiles for a
0.0171 M saline solution flowing through a continuous
microwave heating system at 0.0016 L /s (m easured at the inlet
tem perature)
59
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
TEMPERATURE
( C)
100
O = d a ta a t 0 .0 1 6 L /s
m o d e l e s t im a t e of
p r o file b a s e d o n o u t ­
le t t e m p e r a t u r e
a v e r a g e p o w er a b s o r b e d :
508 W
120
180
240
EXPOSURE TIME (SECONDS)
Figure 16:
30C
360
Experim ental and model based tem perature profiles for a
0.06 M saline solution flowing through a continuous microwave
heating system a t 0.0016 L /s (m easured at the inlet
tem perature)
60
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
1 00
60
V = d a ta a t 0 .0 0 1 6 L /s
O = d a ta a t 0 .0 0 1 6 L /s
m o d e l e s t im a t e of
p r o file b a s e d o n o u t ­
le t t e m p e r a t u r e
40
a v e r a g e p o w er a b s o r b e d :
509 W
20
0
Figure 17:
60
120
180
240
EXPOSURE TIME (SECOND)
300
360
Experim ental and model based tem perature profiles for a
0.12 M saline solution flowing through a continuous microwave
heating system at 0.0016 L /s (m easured at the inlet
tem perature)
61
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
d a t a a t 0 .0 0 1 6 L /s
fo r d is tille d w a te r
d a ta a t 0 .0 0 1 6 L /s
fo r 0 .0 6 M s a lin e
m o d e l e s t im a t e fo r
d is tille d w a te r
m o d e l e s t im a t e fo r
0 .0 6 M s a l i n e
120
180
240
300
360
EXPOSURE TIME (SECONDS)
Figure 18:
Comparison of experim ental and model tem perature profiles
for distilled w ater and 0.06 M saline solution flowing through a
continuous microwave heating system a t 0.0016 L /s (measured
at the inlet tem perature)
62
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
1 00
80
•
= d a t a a t 0 .0 0 2 9 L / s
= m o d e l e s ti m a te of
p ro file b a s e d o n o u t ­
le t te m p e ra tu re
40
a v erag e pow er a b so rb e d :
545 W
20
0
Figure 19:
60
120
240
180
EXPOSURE TIME (SECO ND S)
300
360
Experim ental and model based tem perature profiles for a
0.1 M saline solution flowing through a continuous microwave
heating system at 0.0029 L /s (m easured a t th e inlet
tem perature)
63
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
O = data at 0.0009 L/s
= m o d e l e s t i m a t e of
p r o f ile b a s e d o n
th e o r y and data
a p p ro x im a te pow er ad sorbed
342 W
20
60
Figure 20:
120
150
240
EXPOSURE TIME (SECONDS)
300
3 50
Experim ental and model based tem perature profiles for
distilled w ater flowing through a continuous microwave
heating system at 0.0009 L /s (m easured at the inlet
tem perature)
64
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
( C)
TEMPERATURE
O = d a t a a t 0 .0 0 0 9 L / s
m o d e l e s t i m a t e of
p r o f ile b a s e d o n
th eo ry and data
a p p r o x i m a t e p o w e r a b so r b e d :
362 W
120
180
240
EXPOSURE TIME (SECONDS)
Figure 21:
300
360
Experim ental and model based tem perature profiles for a
0.06 M saline solution flowing through a continuous microwave
heating system a t 0.0009 L /s (m easured at the inlet
tem perature)
65
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
TEMPERATURE(
C)
10 0
80
60
O
= d a ta a t 0 .0 0 0 8 L /s
m o d e l e s t im a t e of
p r o file b a s e d o n
th eo ry and d a ta
4-0
a p p r o x im a te p ow er ab sorb ed :
340 W
0
60
120
180
240
300
360
EXPOSURE TIME (SECONDS;
Figure 22:
Experim ental and model based tem perature profiles for a
0.12 M saline solution flowing through a continuous microwave
heating system at 0.0008 L /s (m easured a t the inlet
tem perature)
66
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
( C)
100
TEM PERA TU RE
SO
60
O = d a ta a t 0 .0 0 0 9 L /s
for d i s t i l l e d w a t e r
A = d a t a a t 0 .0 0 0 9 L / s
fo r 0 .0 6 M s a li n e
40
— = m o d e l e s t i m a t e for
d istilled w ater
- - = m o d e l e s t i m a t e for
0 .0 6 M s a li n e
2 0 Sr
0
60
120
180
240
300
360
EXPOSURE TIME (SECONDS)
Figure 23:
Comparison of experim ental and model based tem perature
profiles for distilled w ater and 0.06 M saline solution flowing
through a continuous microwave heating system at 0.0009 L/s
(measured at the inlet tem perature)
67
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
NON-SOILING REGION
SOILING REGION
POWEIl ADSORBED
(W)
5Q0
450
SALINE CONCENTRATION:
0 .0 6 M
= DATA POINTS AND
MODEL ESTIMATES
FOR POWER UPTAK
4C0
* APPROXIMATE 50ILINCP
POINT
•
0.
w
0 .0 0 0 5
Figure 24:
i
▲
0 .0 0 * 0
X
X
X
0 .0 0 1 5
FLOWRATE ( L / s )
X
X
C.0020
▲
0.002
Power absorbed by a 0.0b to saline solution flowing through a
continuous microwave heating system as a function of flowrate
68
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
50G
NON-BOILING REGION
BOILING REGION
475
POWER ABSORBED
(W)
450
^7*s*
w *
DISTILLED WATER
C = DATA POINTS AND
j
MODEL ESTIMATES
!
FOR POWER UPTAKE
•• = APPROXIMATE BOILING!
POiNT
|
w
p r.p* s
U . W <- J w
0.0020
0.0025
. _ J >» ri.n - — \ - / s ,<
Figure 25:
Power absorbed by distilled water flowing through a
continuous microwave heating system as a function of flowrate
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
FACTOR
DISSIPATION
EFFECTIVE
O = p u r e w ater
a
•
= 0.1 M s a lin e
£.
50
£0
4I uU
f>r*
TEMPERATURE f C )
Figure 26:
Effective dissipation factors for w ater and 0.1 M saline at
2800 MHz as a function c f tem perature
70
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
BAYOL-35
BAYOL-35
EMULSION
EMULSION
EMULSION
SALINE
DISPERSED
UNTREATED
TREATED
EMULSION
EMULSION
EMULSION
Figure 27:
Layer identification for demulsified and undemulsified 0.06 M
saline in Bayol—35 emulsions
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
4
mt
dispersed
phase
•v
50%
w ater recovery
40%
30%
20%
10 %
FRACTION
WATER OR OH, RECOVERED
. 0
WW* * 4
Figure 28:
\j
i
M *U ^
i C t ***\
W
)
Recovery of B ayol-35 and 0.06 M saline from 0.06 M saline in
Bayol—35 emulsions at various dispersed phase concentrations
as a function of coil outlet tem perature
72
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
FRACTION
WATER
OR OIL RECOVERED
oil r e c o v e r y
C.8
% d is p e r s e d
phase
0.6
▼ = 40%
w a te r r e c o v e r y
■ = 30%
▲ = 20%
♦
0.4
= 10%
0.2
A
A
0.0
40
50
60
70
80
MEASURING VESSEL TEMPERATURE ( C)
Figure 29:
Recovery of B ayol-35 and 0.06 M saline from 0.06 M saline in
Bayol-35 emulsions at various dispersed phase concentrations
as a function of initial m easuring vessel tem perature
73
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited without p erm ission.
▼ oil r e c o v e r y
% d isp ersed
phase
o
□
A
FRACTION
f
0.1
-
0.2
-
♦
•
w
■
=
=
=
50%
40%
30%
20%
o
O
-
II
0.6
<
RECOVERED
0.8 -
w ater reco v ery
^ 0 >—
0.001
Figure 30:
O.OO'1
FLOWRATE ( l IT R E S/SE C O N D )
0 .0 0 3
Recovery of Bayol—35 and 0.06 M saline from 0.06 M saline in
Bayol—35 emulsions at various dispersed phase concentrations
as a function of fluid flowrate
74
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
| ”
I
— ■I
T" ~
I' ■
I
* ~l
«
1 p '
»
” I ■
7
80
TEMPERATURE
(°C)
l~
O = DATA
MODEL ESTIMATE
POWER ABSORBED: 405 W
15
w
AXIAL POSITION (m )
Figure 31:
E x p e r im e n ta n d model based tem perature profiles for a 50%
0.06 M saline solution in B ayol-35 emulsion flowing through a
continuous microwave heating system at 0.0018 L /s (measured
at the inlet tem perature)
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
8AYOL-35
EMULSION
EMULSION
EMULSION
SALINE
SALINE
DISPERSED
UNTREATED
TREATED
EMULSION
EMULSION
EMULSION
Figure 32:
Layer identification for demulsified and
Bayol-35 in 0.06 M saline solution emulsions
undemulsified
76
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
□
s a lin e r e c c ^ry
0.8
0
= 37.5%
o
II
S7 . T
*~*
= 50%
B
O . •
.
.1 ,
% d is p e r s e d p h a s e
0.6
□
RECOVERED
GiL OR WATER
FRACTION
---- r—
V
O
O
□
1
3D
-
oo
1 . 0 --------- ----------■----------■--------- ,----------,---------
0.4
0.2
o il r e c o v e r y
•
.
—
■—
,
-
•
*
•
T.
^
0.0 --------- 1-------------------- 1--------- 1----------1----------1----------1----------1----------1--------- 1
80
100
120
COIL CUTLET TEMPERATURE (°C)
Figure 33:
Recovery of Bayol-35 and 0.06 M saline from B ayol-35 in
0.06 M saline emulsions at various dispersed phase
concentrations as a function of coil outlet tem perature
77
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
FRACTION
RECOVERY
OK Oil. OR WATER
’ .0
water recovers
s a lin e c o n c e n t r a t io n
0.A
O
,•
» 0.12 M
V
,T
a 0 .0 0 M
□
, ■ = 0 .0 6 M
u .z
o il r e c o v e r y
-Cw„-4
V V
0 .0
c
J
•L.
CCIL o L /.L c .T i
Figure 34:
□
tn A i
L
A u ( Cy
Recovery of B ayol-35 and saline from 50% Bayol-35 i t saline
emulsions at various saline concentrations as a function of coil
outlet tem perature
78
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
(°c )
TKMI’ERATUKK
80
*r\
60
A?A
MOD
20
POWER ABSORBED: 501 V
A
10
AXIAL LOCATION (rr.)
Figure 35:
Experim ental and model based tem perature profiles for a 507
Bayol-35 in 0.06 M saline emulsion flowing through a
continuous microwave heating system at 0.0025 L /s (measured
a t the inlet tem perature)
79
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
SALINE
SALINE
EMULSION
EMULSION
EMULSION
BITUMEN
D ISPERSED
UNTREATED
TREATED
EMULSION
EMULSION
EMULSION
Figure 36:
Layer identification for demulsified and undemulsified Bitumen
in 0.06 M saline emulsions
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
, V -
29% - 35%
0.8
0.6
RECOVERY
OF
OIL OR WATER
% d is p e r s e d p h a s e
b itu m en recovery
0 .4
FRACTION
o
Oq r
0.2
w ater recovery
0.0
80
Figure 37:
100
FLOWRATE ( L /s )
Recovery of bitum en and 0.06 M saline from bitum en in
0.06 M saline emulsions a t various dispersed
concentrations as a function of coil outlet tem perature
81
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
phase
1 20
TEMPERATURE
( C)
100
80
40
O - DATA
20
= MODEL ESTIMATE
POWER A3S0RBED: 534 W
8
0
io
AXIAL LOCATION (m )
Figure 38:
Experim ental and model baaed tem p eratu re proxies for a 14%
Bitum en in 0.06 M saline emulsion flowing through a
continuous microwave heating system at 0.j019 L/s (measured
a t the inlet tem perature)
82
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
...H ie
PUMP
- 0
- - °
tc
HEATER/TEMPERATURE
CONTROL BATH
>
CONSTANT
TEMPERATURE
BATH
HEATER
HEATER
GLASS COIL
I
I
I
S ch em a tic
sy stem .
d iagram
I
F ig u re 39:
I
I
of
a
c o n v e n tio n a l
co n tin u o u s
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
h ea tin g
7
W fW -
0.8
-
RECOVERED
0.6
-
FRACTION
■ , □ ^ DATA USING THE
MICROWAVE SYSTEM)
0 .4
▼ , V - DATA USING THE
CONSTANT TEMP.
BATH
C o n sta n t
T em perature ^j r
B it u m e n r e c o v e r y
B a th
M icrowave
system
0.2
W ater r e c o v e r y
£j B-ST—
1-6-
0 . 0 L-
80
■I
— I,
I
LI.
...mi
100
90
■ iI
110
COIL OUTLET TEMPERATURE ( C)
Figure 40:
Comparison of bitum en and 0.06 M saline recovery for
microwave heated and conventionally heated emulsions
84
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
120
TEMPERATURE
( C)
00
80
60
40
DATA FROM
MICROWAVE
EXPERIMENTS
20
DATA FROM CONS.
TEMPERATURE
BATH
0
8
4
12
16
AXIAL POSITION ( m )
Figure 41:
Comparison of experim entally based tem perature profiles
between microwave heated and conventionally heated
emulsions of 14% B itum en in 0.06 M saline flowing at
0.0019 L /s (m easured at the inlet tem perature)
85
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
% Water remaining
8C
60
Tap water
40
20
30% w/o emulsion
• Conventional heating
O Microwave heating
20
40
80
60
100
Temperature of water phase (°C)
Figure 42:
Percent w ater rem aining in the emulsion for microwave heated
and conventionally heated 30% w a te r-in —oil(Bayol-35)
emulsions as a function of the tem perature of th e w ater phase
of th e partially demulsified emulsion, (from P al and Masliyah,
1991)
86
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission .
REFERENCES
Chisholm, D., "Predicting Two—Phase Flow Pressure Drop", Encyclopedia of
Fluid Mechanics, vol 3, ch 19, Gulf Publishing Company, USA, 1986.
D atta, K ., "H eat and Mass Transfer in the Microwave Processing of Food",
Chemical Engineering Progress, June 1990, pg. 47-53.
Decareau, R.V., and Peterson, R.A., Microwave Processing and Engineering,
Ellis Horwood Ltd, Chichester, England, 1986.
Fang, C.S., et al, "Microwave Demulsification", Chemical Engineering
Communications, v73, pg. 227—239, 1988.
Fang, C.S., et al, "Oil Recovery and W aste Reduction by Microwave
R adiation", Environm ental Progress, v8, n4, pg. 235—238, November,
1989.
Gerald, C .F., and W heatley, P .O ., Applied Numerical Analysis, 3rd ed.,
Addison—Wesley Publishing Company, Don Mills, O ntario, 1985.
K om olprasert, V. and Ofolio, R.Y., "M athem atical Modeling of Microwave
Heating by Dimensional Analysis", J. of Food Proc. and Pres., 13, 1989,
pg. 87-105
87
o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Kurda, T., Ramaswamy, H.S., Raghavan, G.S.V., and Van de Voort, R.F.,
"Microwave Heating for Pasturization of Milk", American Society of
Agricultural Engineers, Paper No. 893541, 1989.
Lissant, K., Demulsification. Marcel Dekker, New York, USA, 1983.
M eredith, R.E., and Tobias, C.W ., "Resistance to Potential Flow through a
Cubical Array of Spheres", J. of appl. phys., v31, n7, July 1960,
pg. 1270-1273
M etaxas, A.C., "Rapid Heating of Liquid Foodstuffs at 896 MHz", J. of
Microwave Power, 11(2), 1976, pg. 105—115.
M etaxas, A.C., and Meredith, R .J., Industrial Microwave H eating. Peter
Peregrinus Ltd., London, England, 1983.
Ohlsson, T., Bengtsson, N.E., and Risman, P .O ., "The Frequency and
Tem perature Dependance of Dielectric Food D ata as determ ined by a
C avity Pertubation Technique", J. of Microwave Power, 9(2), 1974,
pg. 129-145.
Ohlsson, T ., and Bengtsson, N.E.,"Dielectric Food D ata for Microwave
Sterilization Processing", J. of Microwave Power, 10(1), 1975,
pg. 93-108.
88
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Ozilgen, S. and Ozilgen, M.,"A Model for Pasturization with Microwaves in a
Tubular Flow Reactor", Enzyme Microb. Technol., 13, 1991,
pg. 419-423.
Pal, R., and Masliyah, J., "Use of Microwaves for Demulsification of
W ater—in—Oil Emulsions", AOSTRA Journal of Research, v 7 n 3,
1991, pg. 155-162.
Rayleigh, Lord., Philos. Mag., 34, 1892, pg. 481
Risman, P.O ., Ohlsson, T ., and Wass, B., "Principles and Models of Power
Density Distribution in Microwave Oven Loads", J. of Microwave Power
E2, 1987, pg. 193-198.
Saunders, S., and Masliyah, J., Characteristics and Demulsification of
Solids—Stabilized Emulsions: A Literature Survey and Sum m ary.
D epartm ent of Chemical Engineering, University of Alberta, 1990.
Schramm, L.L., Emulsions: Fundam entals and Applications in the Petroleum
Industry. American Chemical Society, W ashington DC, USA, 1992.
Singh, B., Redford, K., and Puttagunta, V.R., "A Hydraulic J e t Mining
Technique for Recovering Bitumen from A lberta Oil Sands", Proceedings
of the 5th American W ater Jet Conference, August 29—31, 1989,
pg. 275-284.
89
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Spiegel, M R., Schaum ’s O utline of Theory and Problem s .of C om plex Variables
with an Introduction to Conformational M appin£ and its A p p lications.
SI (m etric) Edition, M cG raw -H ill Book Com pany, M ontreal, 1972.
Tinga, W. R. , and Nelson, S O., "Dielectric Properties o f M aterials for
Microwave Processing - Tabulated", J. of Microwave Power, 8(1), 1973,
pg. 23-63.
T inga,
W
k , Personal Com iiv/r
''■> .. 1901.
Thomas, C., Perl, J.P ., and W asan, D.T., "Complex Dielectric Properties of
Macroemulsions in the Microwave Region", J. of Colloid and Interface
Science, vol 139, no 1, 1990, pg. 1-13.
Von Hippel, A.R., Dielectric M aterials and A pplications. The Technology Press
of M .I.T., USA, 1954.
W ickersheim, K., Sun, M., and Kamal, A , "A Small Microwave E—Field Probe
9
Utilizing Fiberoptic Therm om etry", J. of Microwave Power E , 25(3),
1990, pg. 141-148.
Wolf, N.O., "Use of Microwave R adiation in Separating Emulsions and
Dispersions of Hydrocarbons and W ater", C anadian P aten t 1205021,
1986.
90
p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission .
Xi, Weiguo, Personal Communications, 1992.
91
R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
2 982 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа