SYSTEME MULTISONDE POUR MESURER LES CARACTERISTIQUES PROPRES AUX BULLES-TAMPONS INTERDISANT LES DEBITS DE GAZ, DE LIQUIDE ET DE SOLIDE DANS UN LIT FLUIDISE A TROIS PHASES

MULTI-PROBE SYSTEM FOR MEASURING BUBBLE CHARACTERISTICS GAS HOLD-UP, LIQUID HOLD-UP AND SOLID HOLD-UP IN A THREE-PHASE FLUIDIZED BED

Abrégé anglais

32
ABSTRACT OF THE DISCLOSURE
Method and apparatus are disclosed for determining
one or more physical characteristics of individual
bubbles in a gas-liquid system and a gas-liquid-solid
system at high temperatures and pressures. An in situ
probe device is inserted into the system over which
individual bubbles flow. The probe device has a
plurality of independent probes. Each has a rounded
fibre optic end portion projecting into the system. A
source of incident light is directed onto each of the
probes. The rounded end portion of each probe is formed
with a radius of curvature sufficiently large whereby
the angle of incidence of the source light at the
rounded portion is greater than the angle of total
reflection for the fibre optic when in contact with the
gas. The angle of incidence is less than the angle of
total reflection for the fibre optic when in contact
with the liquid. The plurality of probes are spatially
arranged to detect one or more of the bubble physical
characteristics as a bubble flows over the probe device.
The change in light intensity of reflected light
emerging from the probe is measured. The change in
light intensities of each of the probes over time is
evaluated to determine the one or more bubble
characteristics. Each probe is formed of sufficiently
thin fibre optic and spaced from the other probes of the
device to enable detection of the bubble characteristics
for individual bubbles flowing over the probe device.
This probe system enables the monitoring of physical
characteristics of bubbles in two and three phase
systems in an efficient, reliable, economical manner.
The system also provides for a measure of solid, gas and
liquid hold-ups in a three-phase system.

Revendications

24
THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method for determining one or more physical
characteristics of individual bubbles in a gas-liquid
system by use of an in situ probe device inserted into
said system over which said individual bubbles flow,
said probe device having a plurality of independent
probes, each having a rounded fibre optic end portion
projecting into said system, directing a source of
incident light onto each of said probes of said device,
said rounded end portion being formed with a radius of
curvature sufficiently large whereby the angle of
incidence of said source light at said rounded portion
is greater than the angle of total reflection for said
fibre optic when in contact with said gas and the angle
of incidence is less than the angle of total reflection
for said fibre optic when in contact with said liquid,
spatially arranging said plurality of probes of said
probe device to detect one or more of said bubble
physical characteristics as a bubble flows over said
probe device, measuring the change in light intensity of
reflected light emerging from each said probe,
evaluating said change in light intensities of each said
probe over time to determine said one or more bubble
characteristics, each said probe being formed of
sufficiently thin fibre optic and spaced from said other
probes of said device to enable detection of said bubble
characteristics for individual bubbles flowing over said
probe device.
2. A method of claim 1, wherein said probe device is
adapted by way of a particular spatial arrangement of
said probes to determine bubble velocity and cord length
of an individual bubble as it flows over said probe
device.
3. A method of claim 2, wherein said gas liquid system
is fluidized with said bubbles.

4. A method of claim 1 further characterized in
determining one or more bubble characteristics in a
three phase gas-liquid-solid particulate system which is
fluidized by said bubbles.
5. A method of claim 4 adapted to determine said
bubble physical characteristics in a petroleum
hydrocracking system, a coal liquefaction reactor and a
Fischer-Tropsch reactor.
6. A method of claim 5, wherein each of said probes is
formed of a material having an index of refraction
greater than 1.6 for a particular wavelength of said
source of light.
7. A method of claim 5, wherein each of said probes is
formed of a flint glass.
8. A method of claim 1, wherein each said probe is a
fibre optic rod having a rounded tip which is located in
said system, said rounded tip for said fibre optic rod
being determined in accordance with the formulas:
sin.alpha. = ? (i)
and
sin.beta. = ? (ii)
wherein .alpha. and .beta. are the total reflection angles for
gas and liquid respectively, R' is the radius of
curvature of the tip of the probe, a and b are two
characteristic positions in the probe.
9. A method of claim 1, wherein each said probe is
circular in cross-section fibre optic material which is
formed into a U-shaped tip, locating said tip in said
system, the shape of said tip being determined in
accordance with the formulas:
sin .alpha. = <IMG> (iii)

26
sin .beta. = <IMG> (iV)
wherein R is the radius of curvature of said U-shape
tip, d is the fibre optic diameter, x is a
characteristic position of the fibre cross-section,
.alpha. is the total reflection angle for said gas and .beta. is the
total reflection angle for said liquid.
10. A method of claim 4 or 5, wherein particle hold-up
in said three phase system is determined by said probe
device, establishing a base line value for measured
light intensity emerging from each said probe while said
probe is in contact with said liquid, detecting a slight
increase relative to said base line value in intensity
of said reflected light emerging from said probe,
converting said detected slight increase into a value
for increased concentration of said particulate solids
in said system adjacent said probe in accordance with a
predetermined scheme which relates particulate solids
concentration to a detected increase in said light
intensity relative to said base line value of light
intensity of said emerging light.
11. A method of claim 1, wherein said probes of said
device are spatially arranged to provide a first set of
at least two probes spaced apart a predetermined
distance and extending in the direction of flow of said
bubbles over said probe device; a second set of at least
two additional probes spaced apart a predetermined
distance and arranged to extend transversely of said
direction in which said first set of probes extend,
arranging said probes of said first and second sets to
be spaced apart in a sufficiently compact manner to
enable a determination of said physical characteristics,
determining individual bubble velocity by measuring a
first period of time that said individual bubble takes
in flowing over said predetermined distance and
calculating bubble velocity determining cord length of
said individual bubble by measuring a second period of

27
time, that said individual bubble takes in flowing over
one of said probes and calculating cord length from the
product of said calculated velocity and said measured
second period of time, determining bubble cap shape by
comparing relative times when said emerging light
intensity significantly changes and calculating bubble
cap shape from said comparison based on a predetermined
spatial arrangement for said second set of probes.
12. A method of claim 1, further characterized in
directing a source of light from a laser of
predetermined wavelength onto each of said probes and
measuring intensity of said reflected light emerging
from said corresponding probe with a photodetector.
13. A method of claim 12, wherein said laser is a Ne/He
laser emitting monochromatic light with a wavelength of
0.632 microns.
14. A method of claim 13, wherein said probes are
formed of flint glass optic fibre, the fibre having a
diameter of approximately 1 to 3 mm.
15. A method of claim 11 further comprising determining
the duration of change in light intensity emerging from
said probe due to a bubble passing over said probe tip,
making several of said determinations of duration for
bubble passage over said probe tip, converting said
several determinations made during a measured period of
time into a value for hold-up of the gas phase G in
accordance with the formula:
<IMG>
wherein .epsilon. G is the hold-up of the gas phase;
.DELTA. ti is the duration for the ith bubble to
pass over the probe tip; and

28
T is the period during which all of the
bubbles 1 through n were detected by the upper
probe.
16. A method of claim 15, wherein hold-up of liquid is
determined in accordance with a predetermined scheme
represented by the formula:
.epsilon.L = 1 - .epsilon.S - .epsilon.G
wherein .epsilon.L, .epsilon.S and .epsilon.G are values for the hold-up in
the reactor of liquid, solid and gas respectively.
17. An apparatus for determining one or more physical
characteristics of individual bubbles in a gas-liquid
system comprising an in situ probe device adapted for
insertion into such gas-liquid system in an area where
individual bubbles flow in such system, said probe
device comprising a plurality of independent probes,
each of said probes having a rounded fibre optic end
portion, means for directing a source of incident light
onto each of said probes, said rounded end portion
having a radius of curvature sufficiently large whereby
the angle of incidence of said incident light at said
rounded portion is greater than the angle of total
reflection for said fibre optic when in contact with
said gas and the angle of incidence is less than the
angle of total reflection for said fibre optic when in
contact with said liquid, said probes being spatially
arranged relative to each other in a compact fixed
manner to permit detection of one or more of such bubble
physical characteristics as an individual bubble flows
over said probe device, means for detecting a change in
light intensity of reflected light emerging from each
said probe, means for evaluating said change in light
intensities of each said probe over time to determine
said one or more bubble physical characteristics, said
fibre optic of each probe being sufficiently thin to

29
enable detection of said bubble characteristics for
individual bubbles flowing over said probe device.
18. An apparatus of claim 17, wherein each of said
fibre optic probe is formed of a material having an
index of refraction greater than 1.55 for a particular
wavelength of said light source directing means.
19. An apparatus of claim 18, wherein said fibre optic
probe is a flint glass.
20. An apparatus of claim 17, wherein said probe device
includes a body portion with a plurality of apertures
extending therethrough, said apertures extending
generally parallel to each other, the number of said
apertures being equal to the number of said probes, each
of said probes having a protective sheath extending
along said probe leaving said rounded tip exposed, said
sheath being sealingly fixed in a respective said
aperture to position said tip a predetermined distance
outwardly of said body portion.
21. An apparatus of claim 17 or 20, wherein each said
probe is a circular in cross-section fibre optic
material and having a dome-shaped tip in accordance with
the formula:
sin .alpha. = ? (i)
sin .beta. = ? (ii)
wherein .alpha. and .beta. are the total reflection angles for
gas and liquid respectively, R' is the radius of
curvature of the probe, a and b are two characteristic
positions in the probe.
22. An apparatus of claim 17 or 20, wherein each said
probe is a fibre optic rod having a U-shaped tip in
accordance with the formulas:

sin .alpha. = <IMG> (iii)
sin .beta. = <IMG> (iv)
wherein .alpha. and .beta. are the total reflection angles for
gas and liquid respectively, R is the radius of
curvature of the U-shaped probe, d is the fibre optic
diameter and x is a characteristic position of the fibre
cross-section.
23. An apparatus of claim 17, wherein said probes of
each probe device are separated into first and second
sets, said first set of probes comprising at least two
probes spaced apart a predetermined distance and
extending in the direction of flow of said bubbles over
said probe device, said second set of probes comprising
at least two probes spaced apart a predetermined
distance and extending transversely of said direction in
which said first set of probes extend.
24. An apparatus of claim 23, wherein said first set of
probes comprise two probes spaced apart a predetermined
distance less than the average expected cord length of
individual bubbles in the system and second set of
probes comprise two probes each positioned to a side of
an upper probe of said first set, said two probes of
said second set being spaced apart a predetermined
distance less than the average expected width of
individual bubble cap.
25. An apparatus of claim 24, wherein said proves are
formed of flint glass optic fibre, said fibre having a
diameter of approximately 1 to 3 mm.
26. An apparatus of claim 24, further characterized in
first means for measuring a first period of time that an
individual bubble takes in flowing over said
predetermined and means calculating bubble velocity
based on a value for said first period of time measured

31
by said measuring means, said first means measuring a
second period of time that each individual bubble takes
in flowing over one of said probes of said first set and
means for calculating cord length from the product of
said calculated velocity and said measured second period
of time, means for comparing relative times when said
emerging light intensity significantly changes and means
for calculating bubble cap shape from said comparison of
relative times based on said predetermined spatial
arrangement for said second set of probes.
27. An apparatus of claim 26, wherein a laser produces
said source of light.
28. An apparatus of claim 27, wherein said laser is a
He/Ne laser emitting light having a wavelength of 0.632
microns.

Description

12533S7
MULTI-PROBE SYSTE~5 FOR MEASUP~ING BUBBLE CHARACTERISTICS
GAS HOLD-UP, LIQUID HOLD-UP AND SOLID HOLD-UP IN A
THRE~-PHASE FLUIDIZED BED
FIELD OF THE~INVENTION
This invention relates to probes for determining
one or more physical characteristics of individual
bubbles in a gas-liquid phase system or a gas-liquid-
solid particulate phase system.
BACKGROUND OF THE INVENTION
Two- and three-phase systems are often employed in
chemical reactions. In systems involving the use of
reactant gas, the gas is normally contacted with the
liquid or solid by bubbling it through the two or three
phase system. In these systems, particularly with
gas-liquid and gas-liquid-solid particulate phases, the
bubble velocity and bubble size become very important
parameters in determining behavior of the system and its
performanceO Such characteristics are particularly
important with three phase fluidized beds where the gas
and liquid phases act as a media for fluidizing the
solid phase. The physical characteristics of the
bubbles in fluidized beds affect segregation between
reacting phases, bed mixing, particle entrainment from a
dense bed, gas hold-up and solid hold-up, mass transfer
between the gas and the liquid phase and the solid
particulate phase, heat transfer to immersed tube
exchangers and bed expansion. Particulate catalysts are
used in petrochemical hydrocracking processes and with
newly developed very active catalysts, conversion is
limited by the transfer of hydrogen from the bubbles in
the three phase system to the liquid phase.
Presently, unsatisfactory devices are available for
measuring bubble characteristics in three phase
fluidized bed systems. In situ sensors havé been used;
however, they require special design to minimize flow
disturbances. Systems, which require bubble sensors,
tend to operate at high temperatures and pressures; for
example, conditions of heavy oil hydrocracking processes
Fischer-Tropsch reactors and H-coal units. Presently,
there are no commercially available bubble sensors or

2 ~ 33~7
probes which can be used in the extreme environments of
heavy oil hydrocracking or coal liquefaction reactor.
In the process industry, bar level detectors are in
use to control the liquid level in liquid containers.
The bar level detector is made of a glass bar having an
end with a prismatic shape close to 45 angle which,
when it contacts water, beam refraction takes place.
When the end is in water/ no or low light levels are
recorded due to refraction and consequent loss of light
energy into the liquid. A photodetector may be used to
detect the intensity of light reflected back out of the
bar probe. When the bar probe is in the air of the
tank, total reflection of the beam takes place and in
that instance, a significant increase in light intensity
impinging on the photodetector is observed. The changes
of light intensity transformed into voltage variations
in the photodetector provide the information for an
accurate control of liquid tank levels.
Many other types of optical probes are known for
use in measuring changes in composition of two and three
phase systems. Optical fibre probes have been used in a
variety of configurations to detect changes in various
parameters of compositions of two and three phase
systems. An example is disclosed in United States
patent 4,240,747. A fibre optic probe of a diameter of
approximately 1.75 millimeters is formed into various
complex curved shapes. The formed probe is placed in a
liquid. Light emerging from the probe is sensed to
provide information which represents the refractive
index of a liquid. The probe may he also use~ to detect
the presence of gar, hubbl~s in a li~uid where a high
frequency ~luctuation in the si~nal at the output
section of the probe indicates the presence of the gas
bubbles. The high frequency variation of the output
signal is due to the probe being considerably larger
than the individual bubbles in the system, so that many
bubbles pass at one time over the probe causing a rapid
variation in the output signal of light reflected in the
probe.

3 1253357
One of the difficulties with three-phase fluidized
bed systems is to provide a system which can properly
assess local solid hold-ups in the fluidized bed.
Electroconductivity probes have been used which respond
to the difference between liquid, solids and gas
dielectric constants and conductivities. However, this
approach is limited to systems where the appropriate
combination of electrical properties allows the
evaluation of the various hold-ups. The system is not
readily applicable to catalytic hydrocracking of
hydrocarbons or coal liquefaction, because the liquid
phase has a very low electroconductivity.
Gamma absorption techniques have been used to
assess local hold-ups of solids in fluidized bed
systems. The probes used in the gamma absorption
technique cannot be arranged in a multi-probe
configuration with probes spaced one to two centimeters
apart without interference, resulting in the gas liquid
and solid hold-ups not being measured simultaneously.
This system is disclosed in Vasalos, I.A., D.N. Rundell,
K.E. Megirls and G.J. Tjatjopoulos, "Hold-Up
Correlations in Slurry Solid Fluidized Beds", AIChE
Journal, 2~, 2, 346 (1982).
The problem with existing techniques in determining
local hold-ups in fluidized beds is that only average
hold-up values are provided. In an only very limited
situations can local hold-ups be determined by
electroconductivity probes.
SUMMARY OF TH~ INVENTION
Accordin~ to an ~sp~ct of the inventlon, a mekhod
is provided for determining one or more physical
characteristics of individual bubbles in a gas-liquid-
solid system. An in situ probe device is inserted into
the system over which the individual bubbles flow. The
probe device has a plurality of independent probes, each
probe having a rounded fibre optic end portion
projecting into the system. A source of incident light
is directed onto each of the probes of the device.

12S33S'7
The rounded end portion is formed with a radius of
curvature sufficiently large whereby the angle of
incidence of the source light at the rounded portion is
greater than the angle of total reflection for the fibre
optic when in contact with the gas and the angle of
incidence is less than the angle of total reflection for
the fibre optic when in contact with the liquid. The
plurality of probes are spatially arranged to detect one
or more of the bubble physical characteristics as a
bubble flows over the probe device. This is
accomplished by measuring the change in light intensity
of reflected light emerging from each of the probes of
the probe device. The change in light intensities
emerging from each probe is evaluated over time to
determine the one or more bubble characteristics. Each
probe is formed of sufficiently thin fibre optic and
spaced from the other probes of the device to enable
detection of the bubble characteristics for individual
bubbles flowing over the probe device.
According to another aspect of the invention, a
base line value may be established for measured light
intensity emerging from each of the probes of the probe
device while the probe is in contact with li~uid with
normal concentration of particles. A slight change of
the base line value in intensity of the reflected light
emerging from the probe is detected. The detected
slight change in value for light intensity is converted
into a value for a concentration of particulate solids
in the system adjacent the probe in accordance with a
predetermined scheme, which relates particulat~ ol~cls
concentration to a detected increase in light intensity
relative to the base line level obtained for the
condition of a liquid free of particles.
According to a preferred aspect of the invention,
at the same time and because the probes determine the
occurrence of bubbles and non-bubbles (liquid-solids)
contacting the probes, the gas hold-up can be
determined. Once the gas hold-up is known and

~S3;:~S7
considering the above, in providing the solid hold~up,
then the following equation gives the liquid hold-up:
L = 1 ~ ~S - ~
wherein L is the hold-up of the liquid phase, S is the
hold-up of the solid phase and EG is the hold-up of the
gas phase. The fibre optic multi-probe of this
invention provides the local hold-ups of the three
phases (gas-liquid-solid) involved in a three-phase
fludized bed reactor.
According to another aspect of the invention, an
apparatus is provided for determining one or more
physical characteristics of the individual bubbles in a
gas-liquid system. The apparatus comprises an in situ
probe device adapted for insertion into the gas liquid
system in an area where individual bubbles flow in the
system. The probe device comprises a plurality of
independent probes. Each of the probes has a rounded
fibre optic end portion. Means is provided for
directing a source of incident light onto each of the
probes. The rounded end portion has a radius of
curvature sufficiently large whereby the angle of
incidence of the incident light at the rounded portion
is greater than the angle of total reflection for the
Fibre optic when in contact with the gas and the angle
of incidence is less than the angle of total reflection
for the fibre optic when in contact with the liquid.
The probes of the probe device are spatially
arranged relative to each other in a compact ~ixed
manner to permit detection of one or more oE such bubble
physical characteristics as an individual bubble flows
over the probe device. Means detects the change in
light intensity of reflected light emerging from each of
the probes. Means evaluates the change in light
intensit~ of each probe over time to determine the one
or more bubble physical characteristics. The fibre
optic of each probe is sufficiently thin to enable

6 ~2S3357
detection of the bubble characteristic for individual
bubbles flowing over the probe device.
According to a preferred aspect of the invention,
the geometric shape of the rounded end portion of the
probe is calculated with either of the following
formulas:
(a) for the dome-shaped probe, the tip curvature is
defined by:
a
sln ~ = R~ (i)
sin ~ R' tii)
wherein and are the total reflection angles for gas
and liquid respectively, R' is the radius of curvature
of the probe, a and b are two characteristic positions
in the probe.
~b) for the U-shaped probe, the tip curvature is
defined by:
sin ~ = R
R -~ d (iii)
R + x
sin ~ =
R + d (iv)
wherein ~ and ~ are the total reflection angles for gas
and liquid respectively, R is the radius of curvature of
the U-shaped probe, d is the fibre optic diameter and x
is a characteristic position of the fibre cross-section.
BRIEF DESCRIPTION OF THE DRAWINGS
,, _ . . .
Preferred embodiments o:~ the inv~ntion are ~hown .in
the drawincJs wherein:
E'igure 1 is a schematic view of the apparatus
according to this invention for measuring bubble
characteristics;
Figure 2 is a enlarged view of the tip of the probe
of Figure l;
Figures 2a and 2b are enlarged views of the probe
tip of Figure 2 showing details of its geometrical
shape;

7 ~2S~35~
Figure 3 is a schematic representation of an
alternative embodiment for the probe device according to
this invention;
Figure 4 is a plan front view of the support casing
for the individual probes of the probe device;
Figure 5 is a side plan view of the casing for the
probe device of Figure 4;
Figure 6 is an enlarged view of the tip portion of
the probe U-shaped;
Figure 7 is a graph showing the photodetector
output in millivolts versus time for the two vertical
probes of Figure 4 when a bubble flows over the probe
device;
Figures 8 and 9 are graphs showing the output of
the photodetector in millivolts versus time for a bubble
flowing over the outer probes of Figure 4;
Figure 10 is a representative sketch of a bubble
illustrating its cap portion and the parameters for
calculating the cap shape;
Figure 11 is a plot of hubble velocity versus the
cord length of the corresponding bubble;
Figure 12 is a plot of volumetric fraction of
solids in the system versus photodetector output in
millivolts;
Figure 13 is a side elevation of a cylindrical
fluidized reactor bed;
Figure 14 is a section through the reactor of
Figure 13;
Figures 15a and 15b are plane views o~ the probe
device on each si.cle of the reactor a5 shown ~n th~
section of Figure 1~, and
Figure 16 is a plot of photodetector output for
upper and lower probes of the device of Figuxe 15 versus
time.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
An apparatus 10 for detecting bubble
characteristics in a three phase system is schematically
shown in Figure 1. A probe device includes a pl.urality
of probes 12, only one of which is shown as positioned

8 ~Z533~'7
in situ in the system 14. The plurality of probes 12
may be arranged in a format of Figure 4. According to
this embodiment, the system consists of a liquid medium
with gas bubbles 16 and solid particulate matter 18.
The bubbles rising in the direction of arrow 20 form a
fluidized bed for the system 14 within the reactor wall
22. The probe 12 is secured in the reactor wall to
remain fixed at least during the sensing operation. In
accordance with this invention, a plurality of probes 12
are provided in association with a probe device, as
shown more clearly in Figures 3 and 4, to determine
various bubble physical characteristics by the apparatus
10. A laser 24 has its source of light as emitted
therefrom directed along arrow 26 through a beam
splitter 28 onto the exposed end portion 30 of probe 12.
The probe 12 functions as a wave guide to permit light
to travel to the end of the probe and permit ref]ected
light to travel back out of the probe. Incident light,
which is reflected back out of the probe 12 by its
rounded tip portion 32, is split off by beam splitter 28
in the direction of arrow 34. A photodetector 36
detects the intensity of reflected emerging light from
the probe 12. The photodetector 36 is connected to an
analog to digital converter for converting the analog
signals from the photodetector 36 into digital signals
which are input to the microprocessor 40. Programs are
provided in the microprocessor which enable it to
evaluate and analyze the input data to determine
physical characteristics of the bubbles 16 in thc three
phase system 14 oE the reactor.
Various tip configurations may be used for each of
the probes 12. An example of a probe tip 32 is shown in
Figure 2. The probe is encased in a sheath portion 40
which may be stainless steel. The stainless steel
sheath protects the probe from abrasion along its entire
exposed length and also lends structural stability and
impact strength to the probe. The exposed tip 32 i6 a
continuation of the fibre optic extending through the
sheath 4Q. According to the preferred embodiment of

~ZS33S7
Figure 2, the tip end 42 is rounded with a radius of
curvature indicated by arrow 44. The radius of
curvature 44 for the tip end 42 defines a dome-shaped
head which provides the necessary geometric shape for
the interface between the medium of the fibre optic and
that of the system 14, so that the tip can be utilized
to distinguish between solids, liquid and gas bubbles of
the system 14.
The basic principle on which the design and shape
for the tip end 42 is determined requires a radius of
curvature sufficiently large whereby the angle of
incidence of the source light at the rounded portion is
greater than the angle of total reflection for the fibre
optic when in contact with the gas of a bubble and the
angle of incidence is less than the angle of total
reflection for the fibre optic when in contact with the
liquid. With this relationship, a significant fraction
of the incident light xadiation is reflected at the
rounded tip end when the tip is in contact with a gas.
However, when the rounded tip end is in contact with a
liquid, minimal reflection of the incident light occurs
resulting in a significant decrease in reflected
radiation emerging from the fibre optic probe.
The shape of the probe tip may be calculated in
accordance with the following formulas involving two
characteristic angles and which correspond to the
angles of total reflection of the probe immersed in air
and water respectively:
sin (~ = ~l (i)
sin ~ R~ (ii)
wherein R' is the radius of curvature of the probe (44
in Figure 2), a and b are two characteristic positions
in the probe.
For instance if a dome-shaped probe is designed to
be used in a through-phase fluidized bed constituted by

12S3357
~o
air-water-particles, a 2 mm fibre optic bar with R' = 1
mm can be selected.
With reference to Figures 2(a) and 2(b), the above
formulas may be used as fol.lows to calculate probe tip
geometrical shape:
(i) for air:
sin ~ = R~
~ = 43.3 [air-fibre optic(silica cord)~
sin ~= sin 43.3 = 0.6858
R' = 1 mm
a = R'sin ~ = 1 mm x 0.6958 = 0.6858
a = 0.6858
(ii) for water:
sin ~ = R~
~, = 65.8 [water-fibre optic(silica cord)]
sin ~= sin 65.8 = 0.9121
R' = 1 mm
b = R'sin~ = 1 mm x 0.9121 = 0.9121 mm
Based on the above parameters, for air contacting
the tip of the probe,
fraction of the Beam Lost ~r a2 0.68582 0.470
Therefore, fraction of beam intensity lost when air
is contacting the tip i9 O. 470 and fraction of beam
intensity conserved when ~i.x i~ contacting the tip is
0.53.
When water is contacting the tip of the probe,
fraction of Beam Lost ~rb~ 0 91212 0.8319
Therefore, the fraction of beam inter-sity lost when
water is contacting the tip i9 O. 832 and fraction of
beam intensity conserved when water is contacting the
tip is 0.168.

11 ~25335~
In an alternative application of the probe device,
such as, in heavy oil hydrocracking, the following
conditions apply:
heavy oil n = 1.5
hydrogen n = l
glass bar probe
of heavy flint n = 1.6
The following calculation based on the above
formulas ~i) and (ii) may be made:
sin ~= 0.625 = a
R'
a = 0.625 x 1 = 0.625mm
sln~ = 0.9375 = _
15 b = 0.9375 x 1 = 0.9375
For H2 contacting the tip, fraction of beam
intensity lost is:
a2 0 6252 = 0,390,
Therefore, the fraction of beam intensity conserved
is 61%.
For heavy oil contacting the tip, fraction of beam
intensity lost is:
~ b2 _ 0.93752 = 0.878
~r R 2 12
Therefore, the fraction of beam intensity conserved
is 12.2~.
Figure 3 illustrates the dome-shaped probe device
46 mounted in the wall 22 of the fluidized bed reactor.
The probe device 46 comprises a plurality o probes 48
which extend through the probe device suppor~ portlon
50, as mounted in the reactor wall 22. ~he individual
probes 4B have a stainLess steel sheath 52 housi.ng the
individual probes with the exposed tip end 54. Each
probe 48 comprises a wave guide fibre optic 56 which
passes through the stainless steel sheath 52 and is
shaped at its end in the form of a dome. The incident
light from the source travels in the direction of arrow
58 and is split into incident beams 53 and 55. Beam
splitters 28 are used to direct the incident beams 53
and 55 into the body portion 56 of each probe of the

12 12S3;~57
devices. The reflected light radiation emerges from
fibre optic body 56 of each probe and by each beam
splitter 28 is directed into photodetectors 57 and 59 at
90 with respect to the fibre optic to detect the
intensity of the radiation reflected at the dome-shaped
tip 54 of the probe.
As shown in Figure 4, a plurality of the probes 48
are mounted in the support plate 50. The probes, as
housed in their stainless steel sheaths, extend through
the plate 50 in the manner shown in Figure 5. According
to this preferred embodiment, to detect physical
characteristics of the bubbles in terms of bubble
velocity, bubble cord height, bubble cap shape and solid
hold-ups in the fluidized bed, the probes 48 may be
arranged in a T-shaped pattern as shown in Figure 4.
Alternative configurations are useful in this regard
which are discussed in Figures 13 through 15. Probes 48a
and 48b form the first set and probes 48c and 48d form
the second set in defining the T-shaped pattern. The
first set of probes extend in a direction which is in
the same direction as the flow of bubbles over the probe
device 46. Probes 48c and 48d extend transversely of
this direction and spaced to either side of probe 48a,
the purpose of which will be explained with respect to
the analysis of the data, shown in Figures 7, 8 and 9.
An alternate shape for the probe tip, as shown in
an enlarged scale in Figure 6, comprises a U-shaped tip
60 with incident arm 62 and return arm 62a of the wave
guide. A stainless steel sheath 63 is providecl over the
tip. The source oF ~ight enters the probe in ~he
direction of arrow 64 and reflected, emerges in the
direction of arrow 66~ In order to meet the above
qualifications regarding the shape of the fibre optic
probe tip 60, the characteristic angle of the U-shaped
probe, which is the total reflection angle for the probe
in water, and the characteristic angle of the U-shaped
probe which is the total reflection angle for air are
defined in accordance with the following formulas:

13 lZS~3357
sin ~ = R + d (iii)
sin ~ = R + d (iv)
wherein R is the radius of curvature of the U-shaped
probe, d is the fibre optic diameter and x is a
characteristic position of the fibre cross-section as
shown in Figure 6. As an example, in the situation
where ~ and ~ are selected for air and water
respectively, then ~ equals 43.3 for air and equals
65.8 for water. Assuming a fibre optic diameter of .4
mm, R is then equal to 0.87 mm and x is equal to 0.29
mrn. These values indicate that all light rays
positioned on the right side of x = 0.29 mm position as
shcwn in Figure 6 will be lost after the first
reflection. The light rays on the left side of the x
position will be kept inside the fibre after the first
reflection. Comparing x with d, a ratio of 0.725 = x/d
is obtained and one could expect that when the probe is
immersed in water, after the first reflection, 77.6% of
the beam will be refracted and dispersed, while 22.4%
will be conserved in the fibre. The special condition
of the probe design is that a total reflection for the
gas phase occurs without partial refraction for the
liquid phase to provide the necessary perturbations at
the emergent side 66 of the probe.
It is appreciated that the principle of "total
reflection-partial refraction" for a U-shaped probe and
"partial reflection and partial refraction" Eor a
dome-shaped probe can be achi~ved as well in a system
constituted by heavy oil and h~drogen with 1.5 and 1.0
refraction indices respectively. This type of system,
using particulate catalysts, is cornmon to heavy oil
hydrocracking reactors where the probe device, according
to this invention, can be used. With heavy oil systems,
the rnaterial of the probe is selected so as to provide a
refraction index of 1.6. Acceptable materials, which
have a refraction index in excess of 1.6 for the
wavelengths of the light energy to be used, are various

14 ~2~335~
grades of flint glass. Under these conditions, in
accordance with the above formulas, the incident light
will be conserved when the U-shaped probe is contacted
with hydrogen and only 17% of the intensity of the beam
will be transmitted to the emerging arm 66 when the
probe is contacted by heavy oil. For the dome-shaped
fibre, then 61~ of the beam will be conserved when
exposed to hydrogen and only 12.2% when contacted with
heavy oil. It is appreciated that the dome-shaped probe
of Figure 2, being of a solid rod as compared to the
U-shaped probe of Figure 6, has greater mechanical
strength compared to the U-shaped probe. Therefore,
environments where the probes are subjected to physical
abuse, such as mechanical shock abrasion and the like,
the dome-shaped probe is preferred. Such conditions
occur in heavy oil hydrocrackers so that in those
systems, the use of the dome-shaped probe is normal.
From the spatial relationship and known
predetermined spacing between the probes, as shown in
Figure 4, physical characteristics of the bubbles may be
determined such as bubble velocity, bubble cord length
and bubble included angle, that is the shape of the
bubble cap. Typical generally square waves were
recorded when a gas bubble contacts the probe device and
flows over it is shown in Figures 7, 8 and 9. In Figure
7, where the bubble flows upwardly over the probe device
46, the square wave for probe 48b is shown. The bubble
is obviously of a cord length greater than the distance
between the probes 48a and 48b, bec~us~ the squ~rc wave
produced b~ the bubble Elowing over the prob~ 48a
overlaps in real time the square wave for probe 48b.
From the values of Figure 7, the bubble velocity and
bubble cord length may be calculated. Vpon locating on
the graph the starting and ending points of these square
waves for probes 48a and 48b, the mean value estimated
with the first statistical moment is applied to the data
and the duration of each signal may be calculated. The
duration of time for each gradation in Figures 7, 8 and
9 is approximately 0.008 seconds. Normally, only

~25~57
signals coming from probes 48a and A8b are used in
calculating velocity and probe cord length, where such
signals fit the requirement of a standard deviation of
the square waves smaller than 10%. With this constraint
on the data used, virtually all the bubbles selected for
analysis of their physical characteristics travel in a
relatively straight upward manner. The bubble velocity
is calculated knowing the distance between the probes
48a and 48b and dividing it by the separation in the
mean values for the real times of the square waves
representative of the flow of the bubble over the probe
device. The cord length of the bubble may then be
calculated by the product of bubble velocity and the
duration of the square wave signal which indicates the
length of time it takes for an individual bubble to flow
over a probe of the probe device 46.
Another aspect of the multi-probe device, that can
be either consituted by dome-shaped probes or U-shaped
probes, is to determine the shape oE the bubble cap, as
shown in Figure 10 and to assist in providing a careful
screen of the experimental data obtained from the first
set of probes. The shape and size of the bubble may be
reconstructed using the top three probes and comparing
the changes in the signals of the top three probes 48a,
48c and 48d. From this information, the central bubble
cord may be estimated based on the formulas:
L = R-(y - hl)
wherein I, is axial bubble lellgth, R ~ radius curvature
of the spherical bubble cap y is the characteristic
ordinant giving the distance between the central probe
and the centre of the circumference containing the
spherical cap and bl is the ordinant of point 1 of
Figure 10. R may be calculated in accordance with the
formula:
R = (X2 + y2) 0-5

~Z5335~
16
wherein x is the characteristic abscissa giving the
distance between the centre of the circumference
containing the spherical cap and y is as defined above.
The characteristic dimensions of this spherical cap may
then be calculated in accordance with the formulas:
x = [(a4 + b4 )b3 - (b3 + a3 )b4]/~2(a4b3 - a3b4)]
y = (a3 + b32 _ 2a3x)/2b3
wherein al is the abscissa of point 1 of Figure 10; a2
is the abscissa of point 2 of Figure 10; a3 is the
abscissa point 3 of Figure 10 and a4 is the abscissa of
point 4 of Figure 10. Similarly, bl through b4 are the
ordinants of the respective points 1 through 4 of Figure
10. The square wave signals of Figures 8 and 9 thereby
provide the necessary information in calculating the
shape of the physical bubble cap, in accordance with the
above formulas.
The multi-probe device is also useful in assessing
particle hold-up in two-phase fluidized bed (liquid
solid) or three phase fluidized beds (gas-liquid-solid)
involving a liquid and a particulate solid, such as a
catalyst. This system is useful in catal~tic heavy oil
hydrocracking, coal liquefaction reactors and
Fischer-Tropsch fludized units to assess catalyst
hold-up at various levels in the fluidized bed reactor.
By use of the fibre optic probes that withstand the high
pressure and temperatures of these reactors, according
to this invention, the incident light interacts with the
rounded surface wllere the reflected beam is physically
displaced from its point of incidence with the surface.
This effect is a result of a reflected beam crossing the
interface between the probe and the surrounding liquid,
propagating through the liquid, interacting with
adjacent virtual surfaces and re-entering into the fibre
optic slightly displaced. This shlft in the incident
beam is referred to as the Goos-Haenchen effect. With
this shift in the incident beam, the intensity of the

17 ~Z5~33~7
emerging reflected light will vary slightly from a base
llne value representative of the liquid medium free of
particles compared to when particulate catalyst is
adjacent the probe surface. Thereore, over time in
operation of the system, a base line value for the
intensity of the emerging light can be established~ Any
slight increase to a new level in re~lected light
intensity from the base line can be interpreted as being
caused by a change in the fraction of particulate matter
remaining adjacent the probe surface, thereby indicating
a new hold-up of solid particulate matter in that area
of the fluidized bed reactor.
Referring to Figure 12, a plot of the volumetric
fraction of solids versus the base line signal from the
photodetector in millivolts indicates a linear
relationship between an increase in solids adjacent the
probe with corresponding increase in the output signal
of the photodetector. As is apparent from Figures 7
through 9, there is a considerable change in the
photodet0ctor signal when a gas bubble contacts and
flows over one or more of the probes of the probe
device. By establishing a base line value for the
output of the photodetector, when the probe device is
immersed in the liquid and a normal ccncentration of
solids is in liquid about the probe device, then the
solid particle concentration in that region of the
reactor can be measured. A slight change in the base
line value indicates particulate hold-up changes in that
area of the unit. Conversely, a constant hase l~ne
value for the output signal o~ the photodetector
indiccttes that no si~nificant particulate catalyst
hold-up variation takes place in that region of the
fluidized bed reactor. As per Figure 12, the
relationship of volumetric fraction of solids to various
base line output voltages of the photodetector is
provided. The approximately linear relationship between
the volumetric fraction of solids and the base line
level may be used develop a mathematical scheme or
algorithm which can be used to calculate the solid

125335~7
hold-up in the reactor system due to the detected slight
change in output of the photodetector, which may vary
from no output up to 2 milllvolts.
According to another aspect of the invention, aside
from using the probe device to determine hold-up of
particulate catalysts in the region of the probe,
hold-up values for the liquid and gas can also be
determined. The probe device, from which the various
outputs are obtained and analyzed, allows one to
determine the occurrence of bubbles and non-bubbles
contacting the probes. From this information, the gas
hold-up can be determined in providing values for solid
hold-up and gas hold-up. The equation, which provides
the relationship of solid hold-up, gas hold-up and
liquid hold-up, is as follows:
~L = 1 S G
wherein ~L is the hold-up of the liquid phase, S is the
hold-up of the solid phase and ~G is the hold-up of the
gas phase. Substituting the calculated values for S
and G in the above equation, a value for L can be
obtained. Therefore, according to this invention by use
of the probe device, values of the hold-ups of all
phases in the reactor can be calculated to optimize in
the reaction efficiencies of the reactor. This systetn
will be discussed in more detail with respect to Figure
16.
The multi-probe device, according to this
invention, with the plurality of dome-shaped or U-shaped
fibre optic tip portions i; capable of measur1ncJ sev~ral
bubble physical characteristics which, when known, can
result in a more eEfective operation for a gas~ uid or
gas-liquid-solid contacting system. For example, in
three phase fluidized bed systems, bubble velocity,
bubble cap shape, bubble cord length and gas liquid
solid hold-ups provide very useful information in the
control and optimized operation of the fluidized bed
reactor. A further aspect which is considered in the
operation of the fluidized bed reactors is the

il~253;~57
19
relationship of the bubble velocity to the bubble cord
length. As shown in Figure 11, there is a correlation
between the bubble velocity and the cord length which
may be expressed by the formula:
Vb = qL
wherein q and z are coefficients of the bubble velocity
correlation, "L" is the axial bubble cord length and Vb
is the relative velocity of the bubble with respect to
the liquid. For various values of Vl, which is the
superficial liquid velocity in centimeters per second,
it is apparent from Figure 11 that the correlation of
bubble velocity to cord length exists and can be
lS predicted in accordance with the above formula. For the
particular parameters prescribed in Figure 11, q has the
value of 47.08 + 1.04 and z has the value of 0.4302 +
0.05 in the bubble velocity correlation. The following
Table I sets out results obtained for various values of
Vl.

lZS3357
~o
TABLE I
Bubble Velocity Correlation
q and z Parameters
(Vg = 1.214 cm/s)
_ __ _
Number f Vl
RunBubbles cm/s q z
1 7 0.389 39.77 0.2784
10 2 28 0.778 43.75 0.5279
3 23 1.167 47.06 0.413
4 19 1.556 49.03 0.4188
34 1.945 50.61 0.4507
1+2~3+4+5 111 _ 47.08 0.4302
. ~ _ .
As shown in Table 1, the values for q and z are
consistent for various values of Vl to provide thereby a
correlation between the bubble cord length and the
bubble velocity to assist in predicting and controlling
Eluidized bed reactors.
It is appreciated that the manner in which the data
from the probe is analyzed can be accomplished in a
variety of ways. Following the procedures set out with
respect to the system shown in the drawings and using a
computer facility to make the necessary calculations,
according to the embodiment of Figure 1, a hi~h speed
analog to digital converter, which is operated at 2500
samples per second (maximum capacity: 50000
samples/second), is used for digitizing and providing an
input to the microprocessor ~0. A varie~ty o:E
microprocessors may be used such ~s that 801cl by llewlett
Packard under mod~l numbers EIP9826 and HP1000 for
analyzing the input data in accordance with the above
formulas to ascertain the various desired physical
characteristics of the bubbles. By way of programing
the microprocessor, the calculations based on the input
data corresponding measured information from the probes
of the multi-probe device can be made to provide the
information on the desired physical characteristics of
the bubbles.

12S~3S7
2~
sy use of U-shaped probes having fibre optic of
approximately 400 ,um diameter with the radius of
curvature of approximately .5 mm, or dome-shaped probes
made out of 2 mm diameter fibres with a l mm radius of
curvature at the probe tip, the probes may be located to
measure conveniently various bubble sizes. The four
probe devices of the first set used in a column, Figure
4, may be spaced apart 1.25 cm~
An alternative configuration for the probes and
their numbers is shown in Figures 13 through 15. A
cylindrical reactor column 72, as shown in Figure 13,
consists of a plurality of probe support devices 74
located various vertical levels along the height of the
reactor 72. The probe support devices are mounted in
the wall portion 76 of the reactor in the manner shown
in Figure 14. Separate probe support devices 74 are
mounted in each side of the reactor where probe device
78 consists of two probes oriented in the vertical
direction. Probe device 80 consists of three individual
probes oriented in the horizontal direction. ~ith
reference to Figures 15(a) and 15(b), the probe device
80 shows the horizontal orientation for the separate
probes 82, 84 and 86. Similarly, Figure 15(b) shows the
vertical orientiation for the two prohes 88 and 90.
This arran~ement, therefore, provides five probes in the
system where the vertical probes are located in the
column opposite the horizontal set of probes to provide
for an alternative form in measuring the characteristics
of the bubbles and solid and liquid ~as holcl-ups. The
two probes 88 and 90 of probe device 78 are located with
a vertical separatic)n of 0.95 cm. ~he remaining three
probes 82, 84 and 86 of probe device 80 are positioned
at the same level as probe 88. Probes 82 and 86 are
spaced to each side of probe 84 0.95 cm. The outer
probes 82 and 86 extend inwardly further than the
central probe 84 to provide for a 120 circumferential
separation between the central probe 88 distal end and
the distal ends 82 and 86 of the outer probes.

lZS~33S7
~2
This arrangement for the probe devices entering the
reactor column from each side are capable of measuring
axial bubble cord lenght as short as in the range of 0.2
cm.
With reference to Figure 16, the G, the hold-up of
the gas phase may be calculated. Photodetectors are
provided in association with probes 88 and 90 of Figure
15b. A sudden increase in each photodetector output
indicates a bubble having passed over the probe where
each spike in Figure 16 is representative of each square
wave of, for example, the enlarged plot of Figure 7.
Each bubble passes over the probe tip during a finite
time which is in terms of a small fraction of a second
such as in the range of 25 milliseconds to 30
milliseconds. The duration for each bubble to pass over
the probe tip may be represented by t. The gas
hold-up in the system can be calculated in accordance
with the formula:
~n~
li ~ti
G T
wherein G is the hold-up of the gas phase;
ati is the duration for the ith bubble to
pass over the probe tip; and
T is the period during which all of the
bubbles 1 through n were detected by the upper
probe.
Knowing the value for ~S calculated in the manner
previously described, and calculating l`~ as per thc
above, then the hold-up of the liquid phase ll is
calculated as per the formula:
L = 1 - _
From Figure 16, it is also possible to calculate
bubble frequency; i.e., how may bubbles per second
contact the probe tip. For the lower probe, the bubble
frequency is 2.23 bubbles per second whereas the bubble
frequency for the upper probe is 2.30 bubbles per second
.

l~S335~
2~
which provides an average bubble frequency of 2.27
bubbles per second.
Although preferred embodiments of the invention
have been described herein in detail, it will be
understood by those skilled in the art that variations
may be made thereto without departing from the spirit of
the invention or the scope of the appended claims.