Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
1~582~
1 The present invention relates to a mixing
2 apparatus comprising a mixing tank having an off-centered
3 agitator and also preferably employing at least two par-
4 tial baffles which tank is particularly useful for but not
limited to the mixing of buoyant particle slurries.
6 Generally the efficient mixing of buoyant par-
7 ticle slurries is very important to many process opera-
8 tions, as for example in the manufacture of elastomers.
9 In this process buoyant polymer particles in the form of
a water slurry are fed to a slurry drum in the finishing
11 sections and then to product extruders. The primary func-
12 tion of the sluxry drum is to form a uniform slurry and
13 supply it as feed to the extruders. The slurry drum also
14 provides reliable surge capacity between the polymeriza-
tion and solids finishing sections. The mixing of these
16 buoyant particles is also an important operation during
17 other aspects of chemical manufacture, such as during the
18 dissolving of polymer particles in processing lubricating
19 oil additives.
The systems used at present for mixing such
21 buoyant particles slurries generally comprise a baffled
22 tank for receiving the slurries containing means for agi-
23 tating the slurry such as a single or multiple-bladed
24 turbine agitator or stirrer. Ideally, such systems should
produce homogeneous, concentrated particle slurries, the
26 prevention of agglomeration of solids in regions of the
27 tank which could lead to plugging of the tank outlets, and
28 a relatively constant change in the outlet slurry concen~
29 tration to the extruders with changes in the slurry level
in the tank. With the present emphasis on energy conser-
31 vation it is also desired that these systems operate on
32 low powerO
33 Previous attempts to provide efficient mixing
34 systems have included employing unbaffled slurry tanks hav-
ing a centrally disposed agitator to provide a central vor-
36 tex in the tank. However, it has been found that various
37 deficiencies exist in such tanks including the need for
~15~2~
~ 2 --
higher mi~ing speeds and hence more power than normally
anticipated and the concentration of particles between
the bottom of the vortex and top of the vortex-creating
agitator.
Present mixing designs which include baffling
typically have employed full baffles, i.e. baffles which
extend vertically for the full length of the tank walls at
spaced locations and radially a distance of 1/10 or more of
the tank's diameter. This type of full baffling eliminates
tank swirl and vortex formation except when the liquid sur-
face is very close to the top of the agitator. The art has
also disclosed further mixing arrangements wherein baffles
are located at or near the bottom of the tank, at the center
of the tank and annular baffles about the agitator at the
liquid surface.
The combination of partial baffles and a centered
agitator in a mixing system has also been described. In U.S.
Patent No. 4,150,900, issued April 241 1979 to D,L. Smith
Jr, et al, for example, an improved mixing system is des-
cribed which employs partial baffling and a central agi-
tator assembly in a mixing tank. The partial baffles des-
cribed therein comprise small rectangular or triangular
finger baffles which are locatedl at various positions in
the mixing tank. These baffles are secured about the tank
2 5 inner surface and are preferably located just below the
liquid surface level, In the case of variable level mixing
-tanks, fixed partial baffles are employed in the form of
narrow, elongated rectangles located adjacent to the tank
wall and extending from the ma~imum operating liquid level
to the bottom of the tank. If triangular baffles are
employed the size of each can be determined by providing
a horizontal leg which equal~ 10% to 15% of tank diameter
and a vertical leg which equals 15% to 200/o of tank diameter,
If rectangular baffles are used the size of each baffle is
selected so that the sum of the surface areas of the baffles
is the same as the total surface area for four baffles in
a triangular configuration. If four rectangular partial
baffles are used
,
2~
-- 3 --
1 the width of the rectangles in the radial direction should
2 be approxima~ely 2% or 1/50 of the tank diam~ter.
3 In the article, "The Suspension of Floating
4 Solids In Stirred Vessels", G o E ~ H ~ Joosten et. al., Trans.
I. Chem. E. 55, (July, 220, 1977) a number o experiments
6 using baffled and unbaffled vessels equipped with a cen-
7 trally disposed agitator were conducted to determine their
8 effect in forming suspensions of floating particles in a
9 liquid. The optimal flow pattern for suspension of the
floating particles was found to be a fast rotating liquid
11 whose rotation was disturbed by a single baffle having a
12 width equa~ to 0.2 of the diameter of the vessel immersed
13 at the top of the liquid to a wetted depth equal to 0.3
14 of the diameter of the vessel.
The present invention provides a mixing appara-
16 tus comprising a cylindrical tank having a bladed down-
17 pumping turbine agitator assembly associated therewith
18 wherein the shaft of the agitator assembly is displaced
19 radially from the central axis of the tank a distance
equal to from about 1/10 to 1/4 of the tank diameter. By
21 using such an off-centered agitator it has been found that
22 efficient mixing of buoyant particle slurries is achieved
23 with power savings of up to about 38% compared to central-
24 ly or axially disposed agitators.
The apparatus is also equipped with at least two
26 partial baffles and preferably with four partial bafflesl
27 equally-spaced about the inside circumference of the tank.
28 Such par~ial baffles comprise vertical rectangular or tri-
29 angular members attached to the inside wall of the tank
which extend substantially from the bottom of the tank to
31 the maximum slurry height of the tank, each baffle having
3~ a radial width ranging from about 1/36 to 1/100 of the dia-
33 meter of the tank. Particularly preferred are four e-
34 qually-spaced rectangular partial baffles having a radial
width of about 1/50 of the tank diameter. Such a system
36 of partial baffles in conjunction with an off-centered
37 downpumping agitator provide concentrated, homogeneous
38 particle slurries whose characteristics are not appreci-
141 5~2~ -
1 ably changed with changes in the slurry level in the tank.
2 Also, such an apparatus provides excellent dispersion and
3 contacting of systems such as, for example, immiscible
4 liquids, gas and liquids, and gas and slurries.
In the drawings: Figs. 1 and 2 show a side view
6 and a top view of the preferred mixing apparatus of the
7 invention having an off-centered agitator and four equal-
8 ly-spaced partial baffles.
9 Fig. 3 is a top view of an unbaffled mixing ap-
paratus having an off-centered agitator showing, by ref-
11 erence numerals, a set of baffle positions about the cir~
12 cumference of the tank.
13 Referring to Figs. 1 and 2 there is shown a pxe-
14 ferred mixing apparatus 11 according to the invention com-
prising a cylindrical tank 12. An agitator shaft 13 ex-
16 tends axially into the tank and i5 displaced away from the
17 axis of the tank, as shown by the dotted lines, a distance
18 d varying from about 1/10 to 1/4 of the diameter of the
19 tank. The top portion of the agitator 13 is connected
with a means for rotating the shaft (not shown) such as
21 an electric motor. Centrally mounted on the shaft is an
22 agitator assembly 14 shown generally by consisting of 4
23 equally-spaced downpumping turbiIle blades pitched to an
24 angle of 45 as shown and positioned adjacent the bottom
lg of the tank. The bottom section 13a of the agitator
26 shaft may be totally supported in the bottom 19 of the
~7 tank by a steady bearing. Liquid 16 in the tank may be
28 kept at a constant lev~l or may be varied. Spaced 90
29 from one another around the inside wall of the tank are
preferably four (4) rectangular partial baffles 17, each
31 of which comprises a thin, elongated rectangular member
32 comprised of metal, for example extending from the maximum
33 height of the tank to a point just adjacent the bottom of
34 the tank. Each baffle 17 is secured to the lnside wall
of the tank by means of a pair of brackets 18 or other
36 conventional securing means fastened at the top and bottom
37 of each baffle. The radial width of each baffle 17 is
lL 15~2~
-- 5 --
l between about l/36 and l/100 of the width of the tank and
~ preferably about l/50 cf the width of the tank. The baf-
3 fles are spaGed from the wall about 1/50 of the width of
4 the tankO The diameter of the agitator assembly 14 should
be from about l/3 to about l/2 the diameter of the tank.
6 FigO 3 shows a top view of a slurry tank 12 hav-
7 ing off-centered agitator 13 and also indicates 16 possible
g positions, spaced 22.5 apart, for locating one or more
g baffles according to the invention. These positions will
be discussed in more detail in connection with the exam-
ll ples.
12 Experimentally, it has been found that the use
13 of an off-centered agitator in a tank design such as des-
14 cribed above for forming buoyant particle slurries provides
consistently higher outlet slurry concentrations than tanks
16 equipped with a centered agitator. Use of this invention,
17 however, is not restricted to mixing of buoyant particle
18 slurry and has broad utility as mentioned above for immis-
l9 cible liquids, gas and liquid, and gas and slurries and
the like. Moreover, lower tip speeds and hence lower power
21 is required by the of~-centered agitator according to the
22 invention than a centered agitator to achieve the same de-
23 gree of mixing. This savings in power is from about 50%
24 to 80% compared to multiple-turbined central agitators and
fully baffled tanks and up to about 38% in the case of sin-
26 gle turbined agitators.
27 By employing at least two equally spaced partial
28 baffles with an off-centered agitator mixing efficiency
29 is further enhanced. By using four equally-spaced partial
baffles, optimum mixing efficiency is obtained with change~
31 in outlet slurry concentration being relatively small with
32 increases in slurry level.
33 It has also been found that the baffle radial
34 width is important to the mixing efficiency of the inven-
tion. At widths of fxom between l/36 and l/100 of tank
36 diameter, and preferably l/50 of tank diameter, concentra-
37 ted and homogeneous slurries are obtained. At widths sub-
~582~
1 stantially above and helow these ranges the slurry char-
2 acteristics become unsatisfactoryO
3 Triangular bafles may be similarly employed in
4 this invention provided they have surface areas equivalent
to the rectangular partial baffles described above.
6 ~n order to more fully define the present inven-
7 tion, the following examples are given.
8 Example 1
9 In this and the following examples, a 0.9 m dia-
meter tank with a flat bottom was employed comprised of
11 Plexiglas and having a height of 1.8 m. A 0.3 m diameter
12 45 downpumping pitched blade turbine ~ith four (4) 54 mm
13 wide blades was mounted on a vertical agitator shaft dis
14 placed from the axis of the tank by 90 mm (1~10 of tank
diameter). The agitator drive equipment consisted of a
16 3 HP electric motor which, through a variable speed drive r
17 was capable of rotating the agitator at speeds of up to
18 260 RPM. The agitator shaft was supported at the tank
19 bottom by a steady bearing. Four baffles, each 1/50 of
the tank diameter, were mounted at equal intervals inside
21 the internal periphery of the tank spaced from the ~ank
22 wall a distance equal to 1/50 of the tank diameter. This
23 arrangement is shown in Figs. 1 and 2.
24 A Vibrac ~Q-5120 (range 0-5120 in-oz) torque
transducer was included in the shaft for measurement of
26 torque. The TQ-5120 unit was also equipped with a magnet-
27 ic speed transducer which put out 60 pulses per shaft re
28 volution. This was translated to RPM hy m~ans of a fre-
29 quency meter. A slurry outlet, 127 mm above the tank bot-
tom, was connected to the inlet of a positive displacement
31 pump by a 25 mm diameter pipe. The outlet from the pump
32 was directed back into the tank to provide recycle loop.
33 In all the examples the slurry consisted of 4% by weight
34 o 3.2 mm polypropylene pellets in water. The outlet
slurry concentration was measured by collecting samples
3~ at the recycle loop into the tank. The pellets from each
37 sample were filtered and weighed while the volume of water
1 ~58~3~
-- 7 --
1 was measured to determine outlet concentration.
2 In this example a comparison of mixing charac-
3 teristics was made between the off-centexed agitator and
4 a centered agitator used in the same tank having the same
blade pitch and width as the off-centered agitator. The
6 speeds used for both agitator types were 200 and 225 RPM.
7 Outlet slurry concentrations were determined for both agi-
8 tators as the percent of bulk concentration at different
9 slurry levels in the tank measured as the percent (%) of
full height of the tank.
11 The results are given in Table 1 below.
12 Table 1
13
14 ~ _ _ OUTLET SLURRY CONC.(1 of Bulk Concentr_tion
15 ;lurry LevelCentered Off-Centered
(% ofAt Speed - RPM At Speed - RPM
l7 Eull helght) _ 200 225 _200 225
18 48 56.1 65.6 71.3 74.2
37.9 50.3 58.0 61.8
21 As Table 1 shows the o~f-centered agitator pro-
22 vided consistently higher outlet slurry concentrations at
23 each slurry level than those with the centered agitator.
24 In addi~ion, the change in outlet slurry concentration
with change in slurry level was smaller for the off-cen-
26 tered agitator than for the centered agitator.
27 Next, agitator speeds required to achieve an out-
28 let slurry concentration equal to 60% of the bulk concen-
29 tration were then determined for both the centered and off-
centered agitator. It was found that lower speeds were
31 required by an off-centered agitator than a centered agi-
32 tator to achieve the same degree of mixing. The differ-
33 ence in these speeds decreased as the slurry level was in-
34 creased. For the tank design described above the maximum
slurry level was 93~ which c~rresponds to a 14% lower agi-
36 tator speed requirement than the centered agitator system.
37 This can amount to a savings of about 38% in power con-
38 sumption.
~ ~5~234
-- S --
1 Example 2
. _
2 In this Example, outlet slurry concentration
3 measured as the percent of bulk concentration was deter-
4 mined at an 83% slurry level using the tank and baffling
arrangement described in Example 1 except that the agita-
6 tor assembly comprised a 15" diameter pitched blade tur-
7 bine, the agitator shaft of which was displaced a distance
8 equal to 0.225 of the tank diameter. The agitator speeds
9 used were 125 and 150 RPM. Table 2 below summarizes the
10 results.
11 Table 2
12 AGITATOR SPEED I OUTLET SLURRY CONCENTRATION
13 _ (RPM) (% of Bulk Concentratio_) _
14 125 66
16 7~
87 As Table 2 shows high outlet slurry concentra-
tions were obtained at each speed using an agitator cen-
trally offset a distance nearly 1/4 of the tank diameter.
21 Example 3
.
22 Xn this Example, a series of four (4) equally
23 spaced partial baffles as shown :in Fig. 1 were inserted
24 around the inside wall of khe tank at the following posi-
tions, shown in Fig. 3:
26 Series A - Positions 1, 5, 9 and 13
27 Series B - Positions 2, 6, 10 and 14
28 Series C - Positions 3, 7, 11 and 15
29 Series D - Positions 4, 8, 12 and 16
In each series, the baffle width was varied from
31 1/36 to 1/50 to 1/100 of the tank diameter, viz 25mm, 18 mm
32 and 9 mm respectively. Stirring of the off-centered agi-
33 tator was carried out at speeds of 200 and 225 RPM. Out
34 let slurry concentrations measured as the percent of bulk
concentration were determined for each baffle width in each
36 series. The slurry level was 100% of fu~l tank height.
37 Table 3 below summarizes the results.
~ ..
115823~
~o i`l~
V V ~ . .
; ~L
~3 u~ o
,_ ,~ ~
23~
-- 10 --
1 As Table 3 shows, baffle widths from 1/36 to
2 1/100 of tank diameter in conjunction with an off-centered
3 agitator provide efficient mixing. However a baffle width
4 of 1/50 of tank diameter is particularly preferred since
it provided consistently higher slurry outlet concentra-
6 tions at the speeds employed. The position of the baffles
7 did not significantly vary the outlet concentrations.
8 Example 4
9 In this Example the effect of slurry level on
the concentration of outlet slurry was measured using two
11 and four partial baffles with an off-centered agitator as
12 shown in Fig. 1 having radial widths of 1/50 of tank dia-
13 meter. For each baffle arrangement, the slurry level was
14 raised to 28%, 46% and 100~ of full height while outlet
lS slurry concentrations were measured at agitator speeds of
16 200 and 225 RPM. Table 4 summarizes the results.
17
18
21
22
23
24
26
27
28
29
31
32
33
34
36
37
.
11 15~34
~ ~1
~ w ~ ~r ~ ~
.
~.
. . .
1 ~58~
- 12 -
1 With two baffles the outlet slurry concentration
2 increases ~rom about 20% to 90% of the bulk concentra~ion
3 as the slurry level is increased from about 30~ to 100%.
4 With four baffles the outlet slurry concentration decreases
from about 75% to 60~ over the same slurry level rangeO
6 This indicates that a four baffle arrangement is prefer-
7 able to a two baffle arrangement for buoyant particle
8 slurries because of the relatively small change in outlet
9 slurry concentration with varying slurry levels.
11
12
13
14
16
17
18
19
21
22
23
24
26
27
28
29
31
32
33
34
36
37
