Note: Descriptions are shown in the official language in which they were submitted.
;97QO
IMæROVEMENTS IN OR RELATING TO THE THERMAL TREATMENT OF GLASS
This invention relates to the thermal treatment of
glass, and more especially to the thermal toughening of glass
articles, for example flat glass or bent glass sheets. Such
thermally toughened glass sheets may be for use singly as a
motor vehicle windscreen, or as part of a laminated motor
vehicle windscreen, a side light or rear light for a motor
vehicle, or for use in the construction of windscreen assemblies
for aircraft and railway locomotives, or in the construction of
windows for ships, or for architectural uses. Other glass
articles such as pressed or blown glass articles may be therm-
ally toughened by the method of the invention.
The ultimate tensile strength of a glass article can
be increased by a thermal toughening process in which the glass
is heated to a temperature approaching its softening point,
followed by rapid chilling of the glass surfaces to induce
centre-to-surface temperature gradients through the thickness
of the glass. These temperature gradients are maintained as
the glass is cooled through its strain point. This results in
compressive stress in the surface layers of the glass sheet
with compensating tensile stress in the central core of the
thickness of the glass sheet.
Usually this thermal toughening process is carried out
using chilling air directed uniformly at both surfaces of the
glass sheet but it is difficult to obtain a high degree of
toughening using air flows, particularly when toughening glass
sheets of 3 mm thickness or less. Attempts to increase the de-
gree of toughening of a glass sheet by increasing the rate of
flow of cooling air can give rise to loss of optical quality
of the surfaces of the glass and distortion of the shape of the
--2--
- ' ' 10~9~
glass sheet due to the buffeting action of the chilling air.
In another thermal toughening process a glass sheet
at a temperature near to ïts softening poïnt is quenched in a
chilling liquid. High stresses can be produced by this method.
The glass sheets have to be cleaned after quenchïng.
Thermal toughenïng of a glass sheet has also been
proposed by a method in which a hot glass sheet is immersed in
what, in practice, was a freely-bubbling fluidised bed of solid
particles, for example sand.
Such a process has not been brought into commercial
use hitherto.
The major problem which we have found when attempting
to operate such a bed for the thermal toughening of glass is
the high incidence of fracture of the glass sheets during their
treatment in the fluidised bed. The fracture of a glass sheet
while being quenched in a freely-bubbling fluidised bed is
thought to be caused by the induction of destructive tensile
stresses in the leading edge of the glass sheet due to non-
uniform cooling as the leading edge enters the bed of particles
in a state of bubbling or aggregative fluidisation.
Loss of glass sheets due to fracture is particularly
serious when attempting to toughen thin sheets of gJass, for
example of thickness from 2.3 mm to 4.0 mm, to a high stress
value, and has been such as to render the process unacceptable
for the commercial production of toughened glass sheets for
use in car windscreens for example. The problem of fracture
also arises to a lesser but still commercially significant
extent when seeking to toughen thicker sheets, for example up
--3--
106~700
to 8 mm thick.
A freely-bubbling bed in a state of aggregative
fluidisation has also been found to damage hot glass sheets
immersed in it. This is due to the irregular forces to which
the glass is subjected in a freely-bubbling bed. This can give
rise both to changes of overall shape and to more localised
surface damage, the former occurring particularly with thinner
glass sheets such as those of 2 mm to 3 mm thickness. Such
damage as changes of shape may give rise to difficulties in
lamination, and surface damage may give rise to unacceptable
optical quality when the sheet is used as a window or as a
component of a laminated window.
The present invention is based on the discovery that
the use of a gas-fluidised bed in a quiescent uniformly expanded
state of particulate fluidisation unexpectedly produces adequate
stresses in glass sheets quenched in it and substantially reduces
loss of glass sheets due to fracture in the bed or to change
of shape or surface damage so that a successful commercial
yield is achieved.
According to the invention there is provided a
method of thermally treating glass in which the glass is con-
tacted with a gas-fluidised particulate material which is in a
quiescent uniformly expanded state of particulate fluidisation,
to effect heat transfer between the surfaces of the glass and
the fluidised material.
The invention further provides a method of thermally
treating glass, comprising heating the glass to a temperature
above its strain point and immersing the glass in a gas-fluid-
ised bed of particulate material which prior to said immersion
10697Q0
is in a quiescent uniformly expanded state of particulate
fluidisation.
The inventIon is particularly concerned with the
thermal toughening of glass sheets and provides a method of
thermally toughening a glass s-heet comprising heating the
glass sheet and then lowering the hot glass sheet into the
quiescent uniformly expanded bed of particulate material.
Preferably the bed is maintained at a temperature in the range
30C and 150C. This temperature is selected in dependence on
the fluidisation characteristics of the particles and the
required level of stress in the toughened sheets.
The fluidised bed of particulate material in a
quiescent uniformly expanded state of particulate fluidisation,
which is employed in carrying out the invention, can be defined
in terms of the velocity of gas flow through the bed and the
expanded height of the bed. The quiescent uniformly expanded
state of particulate fluidisation exists between a lower limit
of gas velocity at incipient fluidisation, that is the velocity
at which the particles just become suspended in the uniformly
distributed upwardly flowing gas, and an upper limit of gas
velocity at which maximum expansion of the bed occurs while
maintaining a free surface at the top of the bed.
The upper limit of fluidisation gas velocity may
exceed by a small amount the velocity at which the first
clearly recognisable bubble, for example 5 mm in diameter, is
seen to break the calm surface of the bed. One or two such
bubbles may be visible at the gas velocity.
A higher gas velocity results in the development of
extensive bubbling in the bed and at the onset of such bubbling
-5-
- 1065~700
there is partial collapse of the bed height.
We believe that by quenching the sheet in a gas-fluid- -
ised bed which is in a quiescent uniformly expanded state of
particulate fluidisation, any transient tensile stresses in-
duced in the leading edge of the glass sheet on entry into the
fluidised bed are not so severe as to endanger the glass sheet
and to cause it to fracture.
Also the substantially bubble-free nature of the bed
ensures that the hot glass is not subjected to irregular forces
such as could also give rise to fracture, or to changes in shape
of the glass sheet during quenching, or to surface damage.
Previously it has been thougllt that, to obtain a high
heat transfer coefficient between a fluidised bed and an article
immersed in it, it is desirable to maintain a freely bubbling
condition, such that the rapid and continuous movement of the
particles can give rise to transfer of heat between the article
and the bulk of the bed. This, it was thought, would not occur
in a quiescent bed where the particle movement is less. However
it has now been found that unexpectedly high heat transfer
coefficients are obtained between a hot glass article and a
cooler bed of fluidised particulate material in a quiescent
uniformly expanded state and having selected characteristics.
It is found that there is thermal agitation of the
uniformly fluidised particulate material at the hot glass
surfaces when a hot glass sheet is quenched in the bed and there
is greater rapidity of movement and turbulence of the fluidised
particles in the region of the surfaces of the glass sheet than
in the bulk of the bed. This results in a high rate of transfer
of heat away from the glass surfaces. It is thought that
--6--
,
~0697Q(~
particles which become heated by passing in proximity to the
glass surfaces then move rapidly away from the glass sheet and
lose heat to the fluidising air in the bulk of the bed.
A preferred method according to the invention includes
regulating the gas flow to maintain said quiescent state of the
fluidised bed by creating a high pressure drop in the fluidis-
ing gas flow across a membrane through which fluidising gas
enters the bed.
Further according to the invention the particulate
material may comprise particles of density in the range 0.3
g/cm3 to 3.97 g/cm3 and mean particle size in the range 5~m
to l20~m, the material being selected so as to be fluidised in
said uniform quiescent state by fluidising gas flowing uniformly
in the bed at a velocity in the range 0.045 cm/s to 5.61 cm/s.
The density of the particles and their mean particle
sizes are both important in determining the suitability of a
particulate material for consitituting the fluidised bed in a
quiescent uniformly expanded state employed in the method of
the invention. Generally an appropriate particulate material
for fluidisation in a quiescent uniformly expanded state by
fluidising air, when the bed is operating in ambient conditions
of normal room temperature and pressure, is one for which the
numerical product of the particle density, in g/cm3, and the
mean particle size inJlm, does not exceed about 220.
The degree of toughening of a glass sheet which is
achieved by the method of the invention depends on the heat
--7--
1069700
transfer coefficient between the fluidised particulate material
and the hot glass sheet immersed in it. As already described
there is thermal agitation on the hot glass surfaces which give
rise to transfer of heat rapidly away from those surfaces. How-
ever the properties of the particles themselves also affect the
magnitude of the heat transfer coefficient.
For thermally toughening flat soda-lime-silica glass
of thickness in the range 2.3 mm to 12 mm, the method of the in-
vention may comprise heating the glass to a temperature in the
range 610C to 680C, immersing the glass in a fluidised bed in
said quiescent state which has a thermal capacity per unit volume
at minimum fluidisation in the range 0.02 cal/cm3C to 0.37 cal/
cm3C, and maintaining the fluidised bed at a temperature up to
150C to induce in the glass an average central tensile stress
in the range 22 MN/m2 to 115 MN/m2.
The maximum magnitude of average central tensile stress
which can be achieved varies with the thickness of the glass and
the heat transfer coefficient. By selection of a suitable mater-
ial the heat transfer coefficient can be made high enough to prod-
uce toughened glass sheets having a central tensile stress as high
as 40 MN/m2 in glass 2 mm thick, a central tensile stress as high
as 50 MN/m2 in glass 3.0 mm thick, and a central tensile stress as
high as 104 MN/m2 in glass which is 12 mm thickness. However even
higher central tensile stresses than these have been achieved as
is shown in some of the Examples.
The particles may be a non-porous powdered a alumina of
mean particle size in the range 23,~m to 54~um and particle density
3.97 g/cm3, the thermal capacity per unit volume of the bed at min-
imum fluidisation being 0.32 cal/cm3C.
The invention also provides a fluidised bed for use as
a quenching medium for thermally toughening a hot glass sheet,
10697Qo
comprising particles of mean particle size in the range 5~um to
120~um and having a particle density in the range 0.3 g/cm3 to
3.97 g/cm3, and wherein the particles are so selected that the
bed is in a quiescent uniformly expanded state of particulate
fluidisation and has a thermal capacity per unit volume at minimum
fluidisation in the range 0.02 cal/cm3C to 0.37 cal/cm3C.
The invention also comprehends a method of thermally
treating glass in which the glass is contacted with a gas-fluidised
particulate material of non-compacted particle structure which is
such that the apparent density of the particles is less than the
actual density of the material forming the particles and the buoy-
ant particles constitute a gas-fluidised bed in a quiescent unif-
ormly expanded state of particulate fluidisation, the material
forming the particles and the temperature of the bed being so sel-
ected that the heat transfer coefficient of the fluidised bed is
sufficient to produce a desired thermal treatment of the glass as
it cools in the bed.
The invention still further comprehends a method of the-
rmally toughening glass, comprising heating the glass and immersing
the hot glass in a gas-fluidised bed of particles of non-compacted
particle structure which is such that the apparent density of the
particles is less than the actual density of the material forming
the particles and the buoyant particles constitute a gas-fluidised
bed in a quiescent uniformly expanded state of particulate fluid-
isation, the material forming the particles and the temperature
of the bed being so selected that the heat transfer coefficient
of the fluidised bed is sufficient to produce desired toughening
stresses in the glass as it cools in the bed.
The use of particles of non-compacted structure permits
the selection of a material for the particles to give a fluidised
bed having a sufficiently high thermal capacity per unit volume
1069700
at minimum fluidisation to produce a high amount of toughening
stress in the glass whilst avoiding difficulties in fluidisation
of such a material in a quiescent uniformly expanded state of
particulate fluidisation.
The amount of toughening stress produced in the glass
using a fluidised bed comprising particles of a particular non-
compacted material can be controlled by selection of the particle
density. Particles of low density and of a particular size res-
ult in the production of a low amount of toughening stress in the
glass, and the amount of toughening stress produced increases with
increasing particle density up to the maximum density of particles
of such size that they are still fluidised in said quiescent state.
Still further the invention provides a method for therm-
ally toughening a glass sheet, comprising immersing a hot glass
sheet in a fluidised bed in said quiescent state and constituted
by particles of mean particle size in the range 5~m to 120,~m and
apparent particle density in the range 0.3 g/cm3 to 2.35 g/cm3,
the thermal capacity per unit volume of the bed at minimum fluid-
isation being in the range 0.02 cal/cm3C to 0.37 cal/cm3C.
The apparent particle density within a range as set out
above is the actual measured density of the particulate material
taking into account the cavities within the particles, and is to
be distinguished from the true density of the material itself.
By selecting the mean particle size in relation to the
apparent particle density, the suitability of particles of non-
compacted material for constituting the quiescent uniformly ex-
panded fluidised bed can be assessed. Preferably the numerical
value of the product of the apparent particle density, in g/cm3,
and the mean particle size in ~um, should not exceed about 220.
Still further the invention provides a method of ther-
mally toughening glass, comprising heating the glass, and immersing
- 10
10697(~0
the hot glass in a gas-fluidised bed of porous particles in a
quiescent uniformly expanded state of particulate fluidisation,
the material forming the particles and the temperature of the
bed being so selected that the heat transfer coefficient of the
fluidised bed is sufficient to induce desired toughening stresses
in the glass as it cools in the bed.
The particles may be porous particles of ~ alumina of
mean particle size 64~m and apparent particle density 2.2 g/cm ,
the thermal capacity per unit volume of the bed at minimum fluid-
isation being 0.21 cal/cm3C.
In yet another embodiment the particles are of a porous
form of aluminosilicate material of mean particle size in the
range 60~m to 75~m and apparent particle density in the range
1.21 g/cm3 to 1.22 g/cm3, the thermal capacity per unit volume of
the bed at minimum fluidisation being in the range 0.11 cal/cm3C
to 0.19 cal/cm3C.
Further the particles may be of porous powdered nickel
of mean particle size 5~m and apparent particle density 2.35
g/cm3, the thermal capacity per unit volume of the bed at minimum
fluidisation being 0.37 cal/cm3C.
Further according to the invention a method of thermally
toughening glass comprises heating the glass, and immersing the
hot glass in a gas-fluidised bed of hollow particles in a quies-
cent uniformly expanded state of particulate fluidisation, the
material forming the particles and the temperature of the bed
being so selected that the heat transfer coefficient of the fluid-
ised bed is sufficient to induce desired toughening stresses in
the glass as it cools in the bed.
The particles may be hollow glass spheres of mean part-
icle size in the range 77~m to 120~m and apparent particle
density 0.38g/cm3, the thermal capacity per unit volume of the
- 11 -
, ., ~
1069700
bed at minimum fluidisation being in the range 0.05 cal/cm3C to
0.06 cal/cm3C.
In a further embodiment the particles are hollow carbon
spheres of mean particle size 48~m, and apparent particle density
0.3 g/cm3, the thermal capacit-y per unit volume of the bed at
minimum fluidisation being 0.02 cal/cm3C.
The invention also comprehends a fluidised bed for use
as a quenching medium for thermally toughening a hot glass sheet,
comprising particles of non-compacted particle structure which is
such that the apparent density of the particles is less than the
actual density of the material forming the particles, wherein the
mean particle size of the particles is in the range 5~m to 120~um
the apparent particle density is in the range 0.03g/cm3 to 2.35
g/cm3, and wherein the particles are so selected that the bed is
in a quiescent uniformly expanded state of particulate fluidisa-
tion and has a thermal capacity per unit volume of the bed at
minimum fluidisation in the range 0.02 cal/cm3C to 0.37 cal/cm3C.
The invention also includes thermally treated glass
produced by the method of the invention; in particular a thermally
toughened glass sheet produced by the method of the invention.
In order that the invention may be more clearly under-
stood some embodiments thereof will now be described, by way of
example, with reference to the accompanying drawings in which:-
Figure 1 illustrates diagrammatically a verticalsection through apparatus for carrying
out the method of the invention,
/
- 12 -
~.069700
Figure 2 is a detail in section of part of
Figure 1, and
Figure 3 is a graph which illustrates a character-
istic of a gas-fluidised bed in a
quiescent uniformly expanded state of
particulate fluidisation, which is
employed in carrying out the invention.
Referring to Figure 1 of the drawings, a vertical
toughening oven indicated generally at 1 has side walls 2
and a roof 3. The side walls 2 and the roof 3 are made of the
usual refractory material and the bottom of the oven is open,
being defined by an elongated aperture 4 in a baseplate 5
on which the oven 1 is supported. A movable shutter, not
shown, is provided in known manner to close the aperture 4.
A sheet of glass 6 to be bent and subsequently
thermally toughened is suspended in the oven 1 by tongs 7
which engage the upper margin of the sheet 6 and are held
closed in customary manner by the weight of the glass sheet
gripped between the tong points. The tongs 7 are suspended
from a tong bar 8 which is suspended from a conventional hoist,
not shown, and which runs on vertical guide rails 9 which
extend downwardly from the oven to guide the lowering and
raising of the tong bar 8.
A pair of bending dies 10 and 11 are located on either
side of the path of the glass sheet 6 in a chamber 12, which
is heated by hot gas flows through ducts 12a. me interior
of the chamber 12 and the dies 10 and 11 are maintained at the
same temperature as the temperature of the hot glass sheet 6 as
it enters the chamber 12.
- 13
- .
1069700
The die 10 is a solid male die mounted on a ram 13 and
has a curved front face which defines the curvature to be
imposed on the hot glass sheet. The die 11 is a ring frame
female die carried by struts 14 mounted on a backing plate
15 which is mounted on a ram 16. The curvature of the die
frame 11 matches the curvature of the face of the male die 10.
The guide rails 9 extend downwardly through the chamher
12 to either side of the bending dies towards a container for
a gas-fluidised bed 17 of particulate refractory material in
which the hot bent glass sheet is to be quenched. The container
for the fluidised bed comprises an open-topped rectangular tank
18 which is mounted on a scissors-lift platform 19. When the
platform 19 is in its raised position the top edge of the tank
18 is just below the bending dies 10 and 11.
A micro-porous membrane 20, which is described in
greater detail with reference to Figure 2, extends across the
base of the tank 18. The edges of the membrane 20 are fixed
between a flange 21 on the tank and a flange 22 on a plenum
chamber 23 which forms the base of the tank. The flanges and
the edges of the membrane 20 are bolted together as indicated
at 24. A gas inlet duct 25 is connected to the plenum chamber
and fluidising air is supplied to the duct 25 at a regulated
pressure. The membrane is so constructed that fluidising air
flows uniformly into the fluidised bed over the whole base of
the bed to maintain the bed in a quiescent uniformly expanded
state of particulate fluidisation.
Particulate refractory material in the tank 18 is
maintained in the quiescent uniformly expanded state of
-14-
1069~7QO
particulate fluidisation by the upward flow of air uniformly
distributed by the porous membrane 20. The expanded bed is in
a substantially bubble-free quiescent state and there are no
regions of the bed which are not fluidised.
A preferred construction of micro-porous membrane is
shown in Figure 2 and is described in Canadian Application
Serial No. 279,877. This membrane comprises a steel plate 26
which has a regular distribution of holes 27. The margins of
the plate are drilled to provide passages for bolts 24. A
lQ gasket 28 is located between the lower face of the margins of
the plate and the flange 22 on the plenum chamber.
A number of layers 29 of strong micro-porous paper
are laid on the plate 26. For example fifteen sheets of paper
may be used. The membrane is completed with a woven wire
mesh 30, for example stainless steel mesh which is laid on top
of the paper. An upper gasket 31 is located between the margins
of the wire mesh 30 and the flange 21 on the tank.
A basket for catching cullet may be located near the
membrane 20, and is designed so as not to interfere with the
uniform flow of fluidising air upwardly from the membrane.
Referring again to Figure 1, the guide rails 9 extend
downwardly to a position below the bending dies and terminate
in the region of the upper edge of the tank 18. A fixed frame
indicated at 32 is mounted in the tank 18 and has upturned
feet 33 at its base to receive the lower edge of a glass sheet
lowered into the fluidised bed when the tong bar 8 is lowered
beyond the bending dies by the hoist.
With the scissors-lift table 19 lowered and the tongs
7 and tong bar 8 in their lowermost position at the bottom of
`` 1069700
the guides 9, a cool glass sheet to be bent and toughened is
loaded onto the tongs. The hoist then raises the suspended
glass sheet into the oven 1 which is maintained at a temper-
ature, for example 850C, when toughening soda-lime-silica
glass. The glass sheet is rapidly heated to a temperature
near~ts softening point for example a temperature in the
range 610C to 680C.
When the glass sheet has reached a required temperat-
ure uniformly, the shutter closing the aperture 4 is opened and
the hot glass sheet is lowered by the hoist into position
between the open bending dies 10 and 11. The rams 13 and 16
are operated and the dies close t~ bend the sheet. When the
required curvature has been imparted to the sheet the dies
open and the hot bent glass sheet is rapidly lowered into the
fluidised bed in the tank 18 which has been raised to quenching
position by operation of the scissors-lift table 19 while the
glass sheet was being heated in the oven 1.
When high quality laminated glass products are to be
produced incorporating thermally toughened glass sheets
produced by quenching in a fluidised bed an improvement in
optical quality has been observed when the surfaces of the
glass sheet are subjected to a preliminary air cooling just
before the glass is immersed in the fluidised bed. This may
be achieved by locating just above the upper edge of the tank
18 shallow blowing frames which direct cooling air onto the
surfaces of the bent glass sheet as it leaves the bending dies
and enters the fluidised bed.
The preliminary surface cooling is effective to "set-up"
- 16
' ' ' ' ' ' ' U`.~! '
'-' 1069700
the surfaces of the glass sheet and thereby avoid minute
variations in those surfaces such as have sometimes been
observed and which may be due to the thermal agitation of
the fluidised particulate material on the glass surfaces.
Such preliminary surface cooling would however only usually
be employed when the glass is being used for the production
of laminates of high optical quality.
The fluidlsed bed is maintained at a suitable temperat-
ure for inducing a required central tensile stress in the glass, -
for example 30C to 150C, by the water cooling jackets 34 on
the flat longer walls of the tank 18, and by controlling the
temperature of the fluidising air supplied to the plenum ;
chamber 23. The jackets 34 act as a heat sink which absorbs
heat transferred through the bed from the hot glass sheet.
The lower edge of the hot glass sheet is uniformly
ch~lled along its whole length as it enters the horizontal
quiescent surface of the expanded fluidised bed so that there
is no possibility of diffërent tensile stresses being
generated in different areas of the surface of that edge of ~ -
- 20 the glass, such as could lead to fracture. During its descent ~-
. ~'
' ,' ~
- 17 -
700
into the bed the lower edge always contacts fluidised material
in a quiescent uniformly expanded state of particulate fluid-
isation, and this uniform treatment of the lower edge, regard-
less of upward flow of particulate material which may be
generatedcn the hot glass surfaces immediately after they
enter the fluidised bed, largely obviates fracture and the
problems of dealing with glass fragments in the bed. This
together with the avoidance of losses of glass sheets due to
change of shape of the glass sheets and/or damage to the surface
quality, ensures a commercially viable yield of toughened glasses.
Localised thermal agitation of the fluidised bed takes
place on the hot glass surfaces, perhaps by rapid gas expansion
in a manner akin to the boiling of a liquid. The agitation
ensures that there is adequate heat transfer away from the
glass surfaces into the bulk of the fluidised bed, for example
heat transfer coefficients between the bed and the glass sheet
in the range 0.003 cal/cm2C sec to 0.02 cal/cm2C sec are
obtained. The heat transfer continues until well after the
glass has cooled below its strain point, with sufficient
severity to ensure that the centre-to-surface temperature
gradients are maintained as the glass cools through its strain
point, and the toughening stresses are developed thereafter
during the continuous cooling of the glass while it is still
immersed in the bed.
The agitation of the fluidised material at the glass
surfaces sets up currents in the bulk of the bed which ensure
continual dissipation to the remoter parts of the bed of the -
heat which is extracted from the glass by the thermal agitation
-18-
1069700
of the bed in the region immediately surrounding the glass sheet.
The water cooling jackets 34, acting as a heat sink, keep those
remoter parts of the bed cool.
The sheet engages the feet 33 of the frame 32 at the
bottom of its descent, thereby releasing ~he tongs 8. The
glass sheet then rests on the frame 32 while the glass sheet
cools in the fluidised ked. The glass sheet remains in the
fluidised bed until it is cooled sufficiently to be handled
and the tank 18 is lowered by lowering the scissors-lift
platform to expose the fixed frame 32 and the supported
toughened glass sheet which is then removed for subsequent
cooling to room temperature. .
The nature of the quiescent uniformly expanded state
of particulate fluidisation of the fluidised bed is illustrated
in Figure 3 which is a plot of plçnum pressure, that is, the
pressure in the plenum chamber, against the height of the bed
in the tank 18 using ~ alumina particles as described in Example 1 -
2, set out below, and with the tank size and fluidisation
conditions of Example 2, and the temperature of the bed at 80C.
When the plenum pressure reached 15 kN/m2 eY~pansion of ~`
the bed began, the velocity of the fluidising air-through the
bed then being sufficient to produce incipient fluidisation. - ;
That is, at this lower limit of gas velocity the ~ alumina
particles just become suspended in,the upwardly flowing,air_
- Because of the use of a high pressure drop and a
uniformly micro-porous membrane of the kind illustrated in
Figure 2, in which the pressure drop across the membrane is
in excess of 60% of the plenum pressure, there is u~iform
- 19 -
~: . . .... . . . . .
10697~0
distribution of fluidising air flowing upwardly from the upper
face of the membrane. This high pressure drop across the
membrane makes possible sensitive regulation of the velocity
of gas flow upwardly through the particulate material, thereby
permitting regulation of the state of quiescent fluidisation
of the ~ alumina bet~een the minimum fluidisation state just
described and a state of maximum expansion of the bed in which
dense-phase fluidisation is maintained.
This sensitive regulation of the gas velocity is
achieved by regulation of the plenum pressure in the chamber
23, and as the plenum pressure increases there is no sudden
or discontinuous change in the state of the bed. Rather the
- -q~iesce~t uniformly expanded state of the bed persists, as
illustrated in Figure 3, as the plenum pressure is increased
to about 25 kN/m2 and the bed expands to a height of about
102 cm in the tank.
At this plenum pressure the first clearly recognisable
bubble, for example about 5 mm in diameter, may be observed
breaking the surface of the quiescent bed, and this velooity
of the fluidising air may be considered as the minimum bubbling
velocity.
Because of the use of the high-pressure drop membrane -- -
20, it has been possible to observe that this minimum bubbling
velocity is not necessarily the gas velocity producing maximum
- expansion of the bed, and further regulation of the plenum
pressure up to 27 kN/m2 produced a maximum bed height of 105 cm.
While this increase in plenum pressure up to 27 kN/m2 was
effected more small bubbles were observed to break the bed
surface, but the small random bubbles ~Jere not so significant
- 20 -
-
1069700
as to adversely affect the capacity of the bed for quenching
hot sheets of glass, in particular thicker sheets of glass.
With increase of the plenum pressure beyond 27 kN/m2,
persistent bubbling of the bed occured and a tendency of the
bed to collapse to a height below its ma~imum height of 105 cm,
was observed. In this state the bed was unsuitable for tough- ;
ening hot glass sheets.
In this e~amp]e therefore the uniform quiescent expanded
~ state of the fluidised bed of ~ alumina, which was effective
for toughening hot glass sheets is represented by the region of
the curve of Figure 3 lying between plenum pressures of 15 kN/m
and 27 kN/m2, in which region sensitive control of the state of
fluidisation was possible, with consequential control of the
uniform toughening stresses induced in the glass.
The effective heat transfer coefficient of the fluidised
bed relative to the hot glass is determined by the properties
of the fluidising~gas, usually air, the gas velocity in the bed,
the properties of the particulate refractory material notably
the range of sizes of the particles, the mean particle size,
the density of the particles and, in the case when the particles
~ contain cavities, that is have a certain porosity or hollow
- structure, the density of the material of the particles. The --
heat transfer coefficient also depends on the temperatures
of the glass and the bed, since if there is only a small
difference between these temperatures, there will be little
agitation on the surface of the glass and the effective heat
transfer coefficient wi~l be comparatively low.
Other factors affecting the heat transfer coefficient
are the specific heat of the particles, and their average
1069700
heat capacity. In each of the following examples the numerical
value of the product of the particle density, in g/cm3, and
the mean particle size in,~m, is less than 220. This is a
criterion which may be used for assessing the suitability of a
particulate material, that is its capability of being fluidised
by air in a quiescent uniformly expanded state of particulate
fluidisation, operating with ambient conditions of normal
temperature and pressure.
Some examples of toughening of glass ~heets of thick-
ness in the range 2.3 mm to 12 mm, using apparatus as in Figures
1 and 2, and a uniform quiescent expanded bed are set out below.
In each of the following Examples 1 to 11 the edges of the
glass sheet are finished by being rounded using a fine diamond
grit wheel.
Example
~ The particulate refractory is a ~ form of porous alurnina
the properties of which are as follows:- -
Mean particle size (d) = 64~1m
Particle size range = 20 to 160~m
Particle density (~) = 2.2 g/cm3
Material density = 3.97 g/cm3
px d = 14i
Material specific heat = 0.2 cal/gC
Thermal capacity per unit
volume of bed at minimum
fluidisation = ~.21 cal/cm3c
Velocity of fluidising air
in bed = 0.54 cm/s
With the bed maintained at 40C the degree of toughening
of glass sheets of thickness in the range 2.3 mm to 12 mm with
an initial glass temperature in the range 610C to 670C was as
follows:-
- 22 -
1069700
Initial GlassGlass Thickness Average Central
Temperature (mm) Tensile Stress
C ~ , _(MN/m2 )
610 2.3 37
610 - 10 92
610 12 93.5
630 2.3 42.5
630 6 ',72.5
630 12 96
650 2.3 46
650 4 64
650 6 75.7
650 8 92.7
650 10 96
650 12 99
670 2.3 44 ~
670 6 75 '
670 10 100 _
The effective heat transfer coefficient between the
bed and the glass sheets lay in the range 0.01 cal/cm2C sec to
0.012 cal/cm2C sec.
Example 2
In a particular production run using the same ~ form
of porous alumina as in Example 1, bent sheets of glass 2.3 mm
thick were toughened. These sheets were subsequently used as
a component of a laminated windscreen ~r automobiles.
The properties of the ~ alumina are as follows:- I
- 23
1069700
Mean particle size (d) = 64~m
Particle size range = 30 t-o 150,lm
Particle density (~) = 2.2 g/cm3
Material density = 3.9 g/cm
~ d = 141
Size of tank holding 38 cm x 215 cm x
fluidised bed = 105 cm deep
Plenum pressure = 24 kNtm2
Pressure drop across membrane = 15 kN/m2
Pressure drop across membrane = 60% of plenurn
pressure
Xate of flow of fluidising air = 0.175 m3/min
Velocity of fluidising air in
bed = 0.36 cm/s
Temperature of fluidised bed = 60C
Temperature of glass: top edge = 650C to 655C
bottom edge = 670C to 675C
Resulting uniform central
tensile stress in glass = 38MN/m2 to 40MN/m2
The effecti~e heat transfer coefficient between the bed
and the glass sheets lay in the range 0.01 cal/cm2C sec to 0.012
cal/cm2C sec.
Example 3
In another production run, sheets of glass intended as
components of laminated aircraft windscreens and of thickness
3 mm, 4 mm~ 6 mrn, 8mm, and 10 mm, were toughened in a uniform
quiescent expanded fluidised bed of ~ alumina. The same ~ form
of porous alumina was used as in Exarnples 1 and 2.
- ~4
1069700
Size of tank holding
fluidised bed = 45cm x 245cm x 150cm deep
Plenum pressure = 30 kN/m2
Pressure drop across 2
membrane = 19.5 kN/m
Pressure drop across
membrane = 65% of plenum pressure
Rate of flow of
fluidising air = 0.34 m3/min
Velocity of fluidising
air in bed = 0.51 cm/s
Temperature of fluidised O
bed = 60 C
Temperature of glass = 645C to 650C
Resulting uniform centraltensile stress induced in the
glass was as follows:-
Thickness Central Tensile Stress
~ .
3.0 mm 48 MN/m
4.0 mm 53 MN/m
10.0 mm 80 MN/m2
The effective heat transfer coefficient between thebed and the glass sheets lay in thé range 0.01 cal/cm2C sec to
0.012 cal/cm2C ~ec.
Example 4
The particulate refractory material is a porous powdered
aluminosilicate material, each particle containing 13% by -
weight alumina and 86% silica. The powdered material has the
following properties:-
Particle size range = up to 150~m
Mean particle size (d) = 60,um
Particle density (p) = 1.22 g/cm3
- 25
- 10697~)0
Material densi~y = 2.3 g/cm3
fx~ = 73
Material specific heat = 0.38 cal/gC
Thermal capacity per unit
volume of bed at minimum 3
fluidisation = 0.19 cal/cm C
Velocity of fluidising
air in bed = 0.21 cm/s
With the bed maintained a-t 40C, the degree of
toughening of glass sheets of thickness in the range 2.3 mm
to 10 mm was as follows:-
Initial Glass Glass ThicknessAverage Central
Temperature (mm) Tensile Stress
C) (MN/m2)
650 2.3 30.8
650 4 - 44
650 - 6 62.3
650 8 73
650 10 79
The effec-tive heat transfer coe~ficient between the
bed and the glass sheets lay in the range 0.007cal/cm2C sec to
0.009 cal/cm2C sec.
ExamPle 5
Another form of a porous powdered composite alumino-
silicate material was used. Each particle is porous and
- contains 29% by weight alumina and 69% silica. This porous
powder has the following properties:- -
Particle size range = up to 150~m
Mean Particle size (~) = 75,4m
- ~6
10697û0
Particle density (f ) = 1.21 g/cm3
fxd = 91
Material density = 2.3 g/cm3
Material specific heat = 0.2 cal/gC
Thermal capacity per unit
volume of bed at minimum
fluidisation = 0.11 cal/cm3C
Velocity of fluidising air
in bed = 0.33 cm/s
With the bed maintained at 40C, and the initial glass
temperature in the range 610C to 670C, the degree of toughen-
ing of glass sheets of thickness in the range 2.3 mm to 10 mm
was as follows:-
Initial GlassGlass ThicknessAverage Central
l'emperature(mm) Tensile Stress
( C ) (MN~m2 )
610 6
610 10 '74
630 2.3 31.5
630 6 53
650 2.3 33.7
650 4 4~.3
650 6 56
650 8 71.3
650 10 84
670 2.3 32
670 6 58
670 10 81.5
_ ~ . . ~
- 27
~06g7~0
The effective heat transfer coeffi,cient between the
bed and the glass sheets lay in the range 0.007 cal/cm2C sec
to 0.01 cal/cm2C sec.
Example 6
A "Fillite" powder, which comprises the hollow glass
spheres derived from pulverised fuel ash from power station
boilers, was selected to have the following properties:-
Particulate size range = 20 to 160~ m
Mean Particle size (~) = 77 ~ m
Particle density (~) = 0.38 g/cm3
~xd = 29
Material density = 2.6 g/cm3
Material specific heat = 0.18 cal/gC
Thermal capacity per unit
volume of bed at minimum
fluidisation = 0.05 cal/cm3C
the fluidisation velocity
of the air in the
Fillite = 0.11 cm/s
The degree of toughening induced in the glass sheets
which were thermally toughened in this fluidised bed can be
represented by an average central tensile stress which was
measured in conventional manner and the results achieved for
a range of glass thickness from 4 mm to 12 mm, with different
initial glass temperatures in the range 610C to 670C and
with the temperature of the fluidised bed at 40C are as
follows:-
-28-
10697QO
._. ", _ __ -- ___
Ini-tial Glass Glass Thickness Average Cen~ral
Temperature (rnm) Tensile Stress
C ) . (MN/m2 )
610 10 40
610 12 41
630 6 30
630 12 45
650 4 22.4
650 6 32
650 8 37
650 10 39
650 12 48.5
670 6 35
670 10 5o
The effective heat transfer coefficient between the bed
and the glass sheets lay in the range 0.003 cal/cm2C sec to 0.004
cal/cm2C sec.
Example 7 -
Another grade of "Fillite" material was used having the
following properties:-
Mean particle size (~) = 120~ m
Particle density (j) = 0.38 g/cm3
Material density = 2.6 g/cm3
~ = 45
Material specific heat = 0.18 cal/gC
Thermal capacity per unit 30
volume of bed at = 0.06 cal/cm C
minimum fluidisation
Velocity of fluidising
air in bed = 0.27 cm/s
~0697~0
With initial glass tempera-tures in the range 630C to
670C and with the bed at about 40C, stresses induced in
glass sheets of thickness 6 mm to 10 mm were as follows:-
Initial GlassGlass Thickness Average Central
Temperature (mm) Tensile Stress
( C ) (MN/m2 )
630 6 42
630 8 49
650 6 45.5
650 8 51
650 10 63
670 6 48
670 8 . 53
Th.e effective heat transfer coefficient between the bed
and the glass-sheets lay in the range 0.005 to 0.006 cal/cm3C sec.
ExamQle 8
The par-ticulate refractory material used was hollow
carbon spheres of the ki.nd known as "Carbospheres" having the
following properties:-
Particle size range - 5 to 150~m
Mean particle size (~) = 48~lm
Particle density (p) = 0.3 g/cm-'
~x ~ = 14.4 .
Ma-terial density = 2.3 g/cm3
Material specific heat = 0.123 cal/gC
Thermal capacity per unit
volume of bed at minimum
fluidisation = 0.02 cal/cm3C
Velocity of fluidising air
in bed = 0.33 cm/s
~ 30
1069700
The degree of toughening of glass sheets quenched in
this fiuidised bed maintained at about 40C are as follows:-
Initial Glass Glass Thickness Average Central
Temperature (mm) Tensile Stress
C ) . _ (MN/m2 )
610 10 44
630 6 34
650 4 26.3
650 6 32.7
650 8 40
650 10 45
670 6 36
670 10 46
The effective heat transfer coefficient between the bed
and the gl~-ss sheets~lay in the range 0.0035 cal/cm2C sec to 0.004
cal/cm2Csec.
Exam~le 9
The particulate refractory material was porous powdered
nickel having the following properties:-
Mean particle size (~) = 5~um -
Particle density (~) = 2.35 g/cm3
Material density = 8.9 g/cm3
~X~ = 12
Material specific heat = 0.106 cal/gC
Thermal capacity per unit
vol~me of bed minimum
fluidisation state = 0~37 cal/cm3C
Velocity of fluidising
air in bed -- 0~045 cm/s
- 7~1 _
10697VO
~ lass sheets of thickness in the range 2.3 mm to 6 ~m
at an initial temperature of 650C were quenched in a fluidised
bed of this po~us nickel powder which ~as in a quiescent state
and was maintained at about 40C. The degree of toughening
represented by average central tensile stress was as follows:-
Glass Thickness Average Central Tensile
(mm) Stress
(M~/m2)
2.3 77
3 95
_ _ _ 115
_. .
The effective heat transfer coefficient between the bed
and -the glass sheets was 0.02 cal/cm2C sec.
Exan~e _ 10
.
The parliculate material was a non-porous powdered a
alumina. A number of a alumina materials of different mean
particle size were used. All these materials had the following
common properties:-
Particle density (p) = 3.97 g/cm3
Material density = 3.97 g/cm3
Material specific heat = 0.2 cal/gC
The a alumina material was available in different graded
particle sizes of the material and four different fluidised beds
were constituted as follows:-
- 32 -
~0697ao
Alumina Mean Part- Thermal Cap- Thermal Cap- Fluidis-
Bed icle Size pxd acity of acity of mini- ing gas
(d)(~m) Particle mum fluidised ~elocity
(cal/C)(cal/cm3&) (cm/s~
_
A 23 92 5 X 10 9 0.32 1.02
B 29 116 10 X 10 9 0.32 1.62
C 45 180 38 X 10-9 0.32 3.90
¦ D 54 216 66 X 10 0.32 5.61
Glass sheets of thickness in the range 2.3 mm to 12 mm
were quenched in these fluidised beds which are each at a
temperature of 40C. The initial temperature of the glass sheets
was in the range 610C to 670C and the degree of toughening of the
sheets is represented by an average central tensile stress in the
range 42 MN/m to 104 MN/m .
The effective heat transfer coefficient between the bed
and the glass sheets was in the range 0.0062 cal/cm2C sec to
0;0086 cal/cm C sec.
Example 11
A bed of small solid glass spheres was fluidised.
Properties of the bed were as follows:-
Particle size range = 0 to 75 ~m
Mean particle size (~) = 58~ m
Particle density (~1 = 2.5 g/cm3
pxd = 145
Thermal capacity per unit
volume of bed at minimum
fluidisation = 0.34 cal/c3C
Ve]ocity of fluidising
30 air in bed = 0.41 cm/s
.
106970~
Sheets of glass of thickness~ in the range 2.3 mm to
10 mm were heated to an initial temperature in the range
630C to 670C and were quenched in the fluidised bed which
was maintained at a temperature of about 40C.
The degree of toughening of the glass sheets was as
follows:-
Initial GlassGlass Thickness Average Central
Temperature(mm) Tensile S2tress
(C) (~N/m )
630 2.3 38
630 6 72
630 8 87
650 2.3 40
650 6 74.5
650 8 87
650 10 90
670 2.3 43
670 6 80
670 90
The average effective heat transfer coefficient between
the bed and the glass sheets was 0.011 cal/cm2C sec.
To illustrate the high yield of unbroken and undistorted
glass sheets obtained when using a gas-fluidised bed according
to the invention in a quiescent uniformly expanded state of
particulate fluidisation, as compared with the yield when
using a bed in a bubbling state of fluidisation, a number of
similar sheets of glass of size 30 cm x 30 cm and of thickness
2 mm, 6 mm and 12 mm were treated. The glass sheets had an
edge finish in which the edges of the glass sheets were cham-
-34-
697UO
fered using a bonded silicon carbide grinding wheel. This gave
a rougher edge finish than that of the glass sheets of Examples
1 to 11 which were finished with a diamond grit wheel. The
invention made a high yïeld possible even with this rougher,
and cheaper, edge finish.
Each sheet was heated to a temperature as set out below
and then immersed in a fluidised bed of the ~ form of porous
alumina described in Example 1.
For the purpose of these yield tests some hot glass
sheets were immersed in a fluidised bed in a quiescent state
as described in Example 1. A bubbling state of fluidisation
of the bed was then produced by increasing the fluidising
gas velocity above the value producing maximum expansion of the
bed, and an equal number of hot glass sheets were immersed in
the bubbling bed.
The yield of dimensionally acceptable unbroken glass -
sheets, as a percentage of the total number of sheets treated,
was as follows:-
Glass thickness = 2 mm
20Glass Temperature Yield
C QUIESCENT BEDBUBBLING BED
645 95% 52~
660 100% 80%
Glass thickness = 6 mm
.. .
Glass Temperature Yield
C QUIESCENT BEDBUBBLING BED
640 - - -- 80% 40
645 100~ 60
--` 1069700
Glass thickness - 12 mm
Glass Temperature Yield
C QUIESCENT BED BUBBLING BED
635 80% 40%
645 100% 75%
Although the above examples were obtained using 30 cm
x 30 cm square sheets of glass even lower yields with respect
to fracture and distortion result when treating large sheets of
glass such as of motor vehicle windscreen size in a bubbling
fluidised bed. In contrast the yields obtained when treating
such larger sheets of glass in a quiescent fluidised bed are at
least as good as those of the examples referred to above.
The value of the stresses induced in the glass decreases
as the bed temperature increases and in the limit, which may be
about 300C or higher, the stresses in the glass are such that
the glass is annealed rather than toughened. Heating and/or
cooling elements may be provided on the side walls of the tank
18 for controlling the temperature of the fluidised bed. In
all theE~amples the sheets of glass were commercial soda-lime-
silica glass such as is used in the manufacture of aircraft
windscreen panels, automobile windscreens, ship's windows and
architectural panels. Glass of other compositions can be
toughened or annealed in the same way using the method of the
invention. Also articles other than glass sheets, for example
pressed glass articles such as insulators or lens blanks, or
blown glass articles can be toughened or annealed by the method
of the invention.
A fluidised bed according to the invention may be used
for other thermal treatments of glass, for example for the
heating of a relatively cold glass article prior to a further
-36-
ao
processing step, heat transfer from the fluidised material to
the glass which is immersed in the bed being facilitated without
damage to the glass, even when the glass has attained a
temperature at which it is susceptible to damage by irregular
forces.
The invention may also be used for thermally toughening
glass sheets which have been heated and bent while supported
in a near-vertical position, and advanced along a horizontal
path, as described in United Kingdom Patent Application No.
34703/73 (Specification No. 1,442,316). In the apparatus
described in that application the bending dies are enclosed
in a heated chamber which is tilted from an inclined position
to a position in which the bent glass sheet between the bending
dies is vertical and can be lowered vertically into a quiescent
fluidised bed of the kind described above.
In another process employing the invention a glass
sheet may be heated by immersing the sheet in a fluidised bed
which is at a sufficiently high temperature to heat the glass
to pre-bending temperature. After removal from the hot bed
the sheet is bent, and the bent sheet is then toughened by
immersing the glass in a fluidised bed which is in a quiescent
uniformly expanded state of particulate fluidisation as
- described above. The glass sheet could be carried by the same
set of tongs throughout the heating bending and toughening, the
tongs being adjustably mounted so that they move to follow the
bent shape of the glass. In another arrangement each glass sheet
is suspended from non-adjustable tongs for heating and is trans-
ferred to lower-edge support during bending in the manner des-
cribed in United Kingdom Specification No. 1,442,316, the
bent glass sheet being engaged by a second set of tongs which
are arranged according to the bent shape of the glass, and
lowered into the quiescent fluidised bed for quenching.
