Medium Gas flow rate

Beha viors of Formation and Coalescence of Bubbles Injected from a Bottom Porous Nozzle into a. Bath

Masuo Kaji 1, To'oru Sawai 1, and Manabu Iguchi 2

Abstract

Behavior of bubble formation from a bottom porous nozzle in a bath was experimentally investigated for air-water system using a high-speed video camera.

Three bubble dispersion patterns were observed and classfied corresponding to the gas flow rate. Bubbles did not coalesce at low gas flow rate even when the bubbles densely gathered. The transition from the low to the medium gas fLow rate regimes was supposed to occur at a condition where the nozzle outlet superficial gas velocity exceeded the rising velocity of a single small bubble. The transition condition between the medium and the high gas flow rate regimes was not clear. Two-needle nozzles were also tested to investigate bubble coalescence behavior. Three types of bubble formation patterns were observed. Discrete bubbles were formed at low gas flow rate. Coalescence and collapse of bubbles appeared at medium gas flow rate. At high gas flow rate, the front of a long gas jet collapsed and split into small bubbles. Interaction between the two needle nozzles was not clear.

1. Introduction

Gas injection from a nozzle placed at the bottom of a bath is widely applied to metal refining procedures, chemical plants and aeration systems in sewage treatment equipment. The porous nozzle is advantageously utilized to generate small and uniform diameter bubbles in order to enhance mass transfer from gas to liquid and to accelerate chemical reactions in the bath. However, the bubble formation mechanism and the performance of porous nozzle for gas injection are not clarified in detail.

Houghton et al. ⁽¹⁾ investigated the pressure drop in the porous nozzle and observed the shape and dispersion of bubbles in the bath. Koide et al.¹²¹ presented an empirical correlation of average bubble diameter generated from a porous nozzle. Baxter and Wraith^l31 and Goyal et al.¹⁴¹ investigated the effect of gas flow rate on the flow regime of the bubbling jet. They observed discrete bubble columns at low gas flow rate, coalescence of bubbles at medium gas low rate, and bubble blanketing of the entire nozzle outlet at higher gas flow rate. Circulating gas- liquid flow in a bath during gas injection from a bottom porous nozzle was studied by many investigators.(S -⁸⁾ Anagbo and Brimacombe^l91 measured charac- teristics of bubbling jet dispersion, i.e., the radial distributions of void fraction, bubble frequency, mean gas and liquid velocities at various elevations for three gas flow rates of 200, 600 and 1200

cur /

s, which correspond to three different flow regimes observed in ref ( 3)' as shown in Fig. 1 . The authors^UIl investigated

1. Department of Mechanical Engineering, Kinki University, Wakayama, 649-6493 Japan 2. Division of Materials Science and Engineering, Hokkaido University, Sapporo, 060-8628 Japan

2 Memoirs of The School of B.O.S.T. of Kinki University No.4 (998)

the effect of pore diameter and nozzle diameter of the porous nozzle on the bubble dispersion flow regime and presented empirical correlations of the transition boundary between the low and the medium gas flow rate regimes. Nevertheless, the behavior of bubble coalescence has not been revealed.

In this paper, an experimental study is conducted on the behavior of air bubbles injected from a porous nozzle placed on the bottom of a water bath at the atmospheric pressure. Coalescence of bubbles adjacent to the nozzle exit is observed by a high-speed video camera (MEMRECAM Ci/RX- 1) and the transition boundaries between different flow regimes (Fig. 1) is discussed. Fur- thermore, bubble coalescence bevavior is investigated using a two-needle nozzle.

Medium Gas flow rate

High

Figure 1 Illustration of bubble dispersion patterns in three gas flow rate regimes.

2. Experimental apparatus and procedures

Figure 2 shows a schematic diagram of the experimental apparatus. The acrylic cylindrical vessel has an internal diameter of 200 mm and a height of 400 mm. It was filled by water to a depth of 300 mm above the nozzle exit level.

Air was injected through a round porous nozzle into the vessel. Air flow rate is measured by a rota-meter.

The porous nozzle was made of a stainless steel tube with 20 mm in I.D., 24.5 mm in O.D. and 33.5 mm in length as shown in Fig. 3. A porous disk with 20 mm in O.D. and 10 mm in thickness was inserted into the tube and was sealed around the periphery of the porous disk. Three porous disks with average pore diameters of 5, 50 and 400 11m were used in this experiment. The porosity ^E is approximately 0.36 for all the porous nozzles.

The needle nozzle has two needles of 0.18 mm in I.D., 0.52 mm in O.D. and 30 mm in length inserted into a round brass plate as shown in Fig. 3. The distance between the tips of needles were varied as 2.5 and 5 mm.

The nozzle was placed on the inside of the bottom plate of the vessel and the air was injected into the vessel during experiment. Bubble formation and dispersion

of bubbles were observed using a high-speed video camera (MEMRECAM Ci/

RX- 1) at 1000 frames/ s. Video images were analysed by a personal computer using a image processing device.

o

0 o 0

° _°

⁰

00 00

(X)

Video Camera

Control Valve

Air---~

Figure 2 Schematic diagram of the experimental apparatus.

Porous Nozzle Two-Needle Nozzle

Figure 3 Schematic view of the porous and two-needle nozzles.

3. Experimental results and discussion 3.1 Bubble formation from porous nozzle

Figure 4 shows a bubble dispersion flow regime map presented in our prevlOUS paper.uO) The abscissa represents the average pore diameter, d_{pm ,} and the ordinate represents the superficial velocity, ug , which is determined by the volumetric gas flow rate per unit area of the nozzle outlet cross section. A round porous nozzle having a diameter of 20 mm was used in the experiment. Seven kinds of porous disks with various pore diameters were selected to investigate the effect of pore diameter. The bath surface pressure Ps was kept at the atmospheric pressure.

4 Memoirs of The School of B.O.S.T. of Kinki University No. 4 (1998)

P

_s

= 0.1 MPa

d_n

=

^{0.2 m}

e

Vol

0.6

~

=

High gas flow rate

o

0.2

Medium gas flow rate Eq.(I)

______ -"'" __ "__ L ___ _

OL---'--~L...I-L....L.L.L.L---'--~L...I-L....L.L.L.'---'--~.a....&...I...L...LLIO

1 0° 1 0

¹

1 0

²

1 0 3

Pore diameter d

_{pm ,}

lO-6 m

Figure 4 Classification of bubble dispersion patterns into three gas flow rate regimes.

Bubble dispersion regimes were named as "low", "medium" and "high"corre- sponding to the gas flow tate, respectively, as shown in Fig. 1 . In the low gas flow rate regime, discrete bubbles generated from the nozzle were uniformly dispersed without coalescence. As the gas flow rate increased, some of bubbles began to coalesce and gas slugs were formed ("medium" regime). As the gas flow rate increased much more, a large packet which enveloped the whole nozzle outlet was formed and departed from the nozzle surface with a successive bubble formation. In the high gas flow rate regime bubble formation behavior seems to be similar to that of a single hole nozzle.

The critical superficial gas velocity of the transition between the low and medium gas flow rate regimes was reported by Anagbo and Brimacombe(91, Goyal et al. (41, and Baxter and Wraith(31 as 0.14, 0.15 and 0.16 m/ s, respectively. These values agree fairly well with our result presented in Fig. 4. However, for the transition between the medium and the high gas flow rate regimes, they gave different values of the critical velocity as 0.4, 0.25 and 0.67 m/ s, respectively.

This might be caused by the differences in pore diameter, shape of nozzle and observation method. These values roughly agree with our data. In the present study, behaviors of bubbles in the transition region, particularly iri the transition between the low and the medium flow rate regimes, are discussed.

Figure 5 shows typical bubble dispersion flow regimes for the tested three porous nozzles at various gas flow rates. In theses figure, (a) indicates the low, (b) and (c) the medium, and (d) the high regimes, respectively, for each nozzle.

Typical time serial images of bubble formation for the nozzles of dpm = 5 and 50 11m are shown in Figs.6 and 7, respectively. Time intervals from the most left side images are shown below each image.

(b) ug =0.198m1s (c) ug =0.317m/s Cd) ug =0.396m/s dpm =5 11 m(UHE2000)

(b) ug =0. 198m1s (c) ug =0.317m/s (d) u_g =0 396m/s dpm =5011 m (M58)

(b) u_g =0.119m1s (c) u_g =0.476m/s (d) u_g =0.634m/s dpm=400 11 m (PBF12)

Figure 5 Typical examples of still video images showing Ca) low, (b), Cc) medium and Cd) high gas flow rate regimes for different pore diameter nozzles.

6 _{VIemoirs of The} School of B.O.S.T. of Kinki Cni\'efsity NO.4 (998)

Oms 21ms

Oms 19ms

40ms u@=O.198m/s

38ms ug=O.396m/s

80ms 99ms

80ms 99ms

81ms 103ms

Figure 6 Time serial video images of various gas flow rate regimes for the porous nozzle with d_pm = 5 11m.

Oms 20ms 31 ms 43ms 56ms 71ms ug=O.317rn/s <4m=50 ^J1 m

Oms 20ms 40ms 60ms 80ms lOOms

ug=O.396rn/s dpm=50 ^{J1 m}

Figure 7 Time serial video images of various gas flow rate regimes for the porous nozzle with dpm =50 /Jm.

8 Memoirs of The School of B.O.S.T. of Kinki University No.4 (1998)

In the low flow rate regime, the bubble diameter is approximately 3 mm when the pore diameter was less than 100 11m. At the nozzle outlet, a very small bubble catches up a preceding bubble due to the difference in bubble velocity, and the small bubble is absorbed by the preceding large bubble. Bubbles enlarged by such a periodical coalescence procedure have an approximately equal diameter. This coincides with the results of two-needle nozzle experiments as described later.

When the pore diameter is larger than 100 11m, the bubble diameter increases with the increase in pore diameter. Bubbles do not seem to coalesce except in the region immediately adjacent to the nozzle outlet even when bubbles densely gather.

In the medium flow rate regime, bubbles begin to coalesce with each other near the central part of the nozzle outlet as shown in Fig. 5 (b). In the outside part of the nozzle small bubbles are generated. When a large bubble is formed in the central part of the nozzle, the top of the bubble grows up as shown in Figs. 6 and 7 . It is similar to the transient behavior of a single phase flow discharged from a nozzle. The growth velocity of the large bubble (gas slug) is considered to be approximately equal to the nozzle outlet gas velocity. Since the area of gas injection does not occupy the entire surface of the porous plate, the nozzle outlet gas velocity might be greater than the superficial gas velocity, ug , in some part.

On the contrary, the rising velocity of a bubble, whose diameter is larger than approximately 3 mm, in the still water is given by the following empirical equationo^D•

UB

=

1.18 [

a(p,;,

^fo)g

J

^V' ( 1 )

where, a is the surface tension, PI and P_g are the densities of liquid and gas, respectively, and g is the gravitational acceleration. For the present condition

U_B is estimated as 0.19 m/ s.

From the discussion above described, it is considered that the local gas velocity in the central part of the nozzle outlet exceeds the rising velocity of a single small bubble (u_B =0.19 m/ s) when the superficial gas velocity, ug , becomes a certain value, which is slightly less than U B . The transition from the low to the medium regimes may occur under such conditions. This transition condition agree well with the result for d_pm <100 11m as shown in Fig. 4. As the pore diameter increases, inequality of gas velocity distribution across the nozzle outlet is considered to increase and the transition may occur at a smaller value of ug •

Fig. 5 (c) corresponds to the upper limit condition of the transition between the medium and the high gas flow rate regimes. A large bubble envelope almost the entire nozzle outlet and small bubbles are decreased in number. Small bubbles are not absorbed with a large bubble in the region far apart from the nozzle outlet. However, the difference in behavior of the large bubble formation between the medium and the high gas flow rate regimes is not so clear when comparing with Fig. 5 (d).

3.2 Bubble formation from two-needle nozzle

Figure 8 shows examples of bubble formation behavior in the plane domain of 20 x 20 mm above the outlet of two-needle nozzle. When the gas flow rate is low, discrete bubbles are regularly formed. Coalescence and collapse of bubbles are observed at higher gas flow rates. Video images were converted into binary images and the bubble formation characteristics were analyzed using a personal computer.

... _... ... . ..

^II/1II

- ^{., ... ... • ..}

0 - '

-- ^•

1.0 cm³/s 2.8 cm³/s 4.7 cm³/s

12.6 cm³/s

Interval between Nozzles 5.0mm

••

0-

_-.-

_"'

., .

^{"01 •}

-

• Cit

12.6 cm~/s 15.5 cmJ/s Interval between Nozzles 2.5mm

Figure 8 Video images of bubble formation from the needle nozzle at various gas flow rates.

10 Memoirs of The School of B.O.S.T. of Kinki University No. 4 (1998)

Figure 9 shows the relation between the volumetric gas flow rate, Qg, and the bubble frequency,

Ie.

The bubble frequency has a maximum value at a certain value of Qg which is approximately equal to the critical flow rate due to the sonic velocity. In Fig. 10 the average bubble diameter, dB, and the number of bubbles, nB , are plotted against Qg. nB shows the same behavior as the bubble frequency, but the behavior of dB is complicated. The difference. in results of two needle intervals of 2.5 and 5 mm cannot be seen.

100~~~~~~~~~~~~~~

~ N

~

80

~

g

⁶⁰

Q)

&

~ 40

(])

~

~ ~

20

~

Interval between nozzles 5.0mm 0

2.5mm •

•

o

~ 0

• _I

.. I.

0 0

• _{• i}

• •

5 10 15

Gas flow·rate 00 cm

³

/s

Figure 9 Relation between bubble frequency and volumetric gas flow rate for the two-needle nozzle.

Figure 10 Average bubble diameter and number of bubbles plotted against volumetric gas flow rate for the two- needle nozzle.

When the gas flow rate is low (region I in Fig. 10), discrete bubbles are formed periodically. From a detailed observation of the high-speed video images, the bubble formation mechanism was found as follows: Firstly, a bubble grows up to a certain volume and departs from the nozzle outlet. The growth velocity of the next bubble, however, is greater than the rising velocity of the preceding bubble and coalescence of bubbles occur. After the departure of a larger bubble the above described procedures are repeated periodically. The bubble diameter increases with the increase in gas flow rate. The bubbles do not coalesce with each other in the region far apart from the nozzle outlet.

In the region II, a large bubble formed at the needle nozzle outlet is collapsed and split into small bubbles. Although the small bubbles rise in the bath, the needle tip is enveloped by the remaining large bubble., Thus the average diameter of the dispersed bubbles become smaller than that in region 1. The bubble diameter decreases with the increase in Qg as shown in Fig. 10. In the region III, the superficial gas velocity at the nozzle outlet exceeds the sonic velocity and a long gas jet is formed. By the collapse of the gas jet front many bubbles of various sizes are generated. In any region of I, II and III, coalescence of bubbles was not observed in the region far apart from the nozzle outlet. Interaction between the two needle nozzles could· not be clarified.

4. Conclusions

Experiments were conducted on the mechanism of bubble formation from a bottom porous nozzle in a bath for air-water system using a high-speed video camera. Two-needle nozzles were tested to investigate bubble coalescence behavior.

The following conclusions were obtained.

(1) Bubble dispersion patterns were identified as the low, medium and high gas flow rate regimes, respectively, as proposed in our previous paper.oO) In the low gas flow rate regime, bubble coalescence did not occur even when the bubbles gathered densely. The transition from the low to medium gas flow rate regimes occurred when a large bubbles appeared in the central part of the nozzle outlet.

The transition was supposed to occur at the condition where the nozzle outlet superficial gas velocity exceeded the rising velocity of a single small bubble. However, the transition between the medium and high gas flow rate regimes was not clear.

( 2) Three types of bubble formation patterns were observed in the two-needle nozzle experiment. Discrete bubbles were formed at low gas flow rate. Coalescence and collapse of bubbles appeared at medium gas flow rate and the bubble diameter decreased as the gas flow rate increased. At high gas flow rate, superficial gas velocity at the nozzle outlet exceeded the sonic velocity and a long gas jet was formed. The front of gas jet collapsed and split into small bubbles. Interaction between the two needle nozzles was not clarified.

References

( 1) Houghton, G., McLean, A. M. and Ritchie, P. D. (1957) Mechanism of formation of gas bubble-beds, Chern. Eng. Sci., 7, 40-50.

( 2) Koide, K., Kato, S. Tanaka, Y. and Kubota, H. (1968) Bubbles generated from porous plate, J. Chern. Eng. Japan, 1, 51-56.

12 Memoirs of The School of B.O.S.T. of Kinki University No.4 (1998)

( 3) Baxter, R. T. and Wraith, A. E. (1970) Transitions in the bubble formation mode of a submerged porous disc, Chern. Eng. Sci., 25, 1244-1247.

(4) Goyal, P., Themelis, N. J. and Zanchuk, W. A. (1982) Gaseous refining of anode copper, J. Metl., 34, 22-28.

( 5) Johansen, S. T., Robertson, D. G. C., Woje, K. and Engh, T. A. (1988) Fluid dynamics in bubble stirred ladles: Part I, Experiments, Metall. Trans. B, 19B, 745-754.

( 6) Johansen, S. T. and Boysan, F. (1988) Fluid dynamics in bubble stirred ladles : Part II, Mathematical modeling, Metall. Trans. B, 19B, 755-764.

(7) Sahai, Y. and Guthrie, R. 1. L. (1982) Hydrodynamics of gas stirred melts:

Part I, Gas/liquid coupling, Part II, Axisymmetric flows, Metall. Trans. B, 13B, 193-202, 203-211.

(8) Stapurewicz, T. and Themelis, N. J. (1987) Mixing and mass transfer phenomena in bottom-injected gas-liquid reactors, Can. MetaJl. Q., 26, 123-128.

( 9) Anagbo, P. E. and Brimacombe, J. K. (1990) Plume characteristics and liquid circulation in gas injection through a porous plug, Metall. Trans. B, 21B, 637-648.

(10) Iguchi, M., Kaji, M. and Morita, Z., Effects of pore diameter, bath surface pressure and nozzle diameter on the bubble formation from a porous nozzle, Metall. Trans. B, (to be published).

(11) Peebles, F. N. and Garber, H. J. (1953) Studies on the motion of gas bubbles in liquids, Chern. Eng. Prog., 49, 88-97.

3{tj§Pjtd.t 'i~~~~ 7⁰ 0

-e

A ~~ ~~:t ~ ii A ~ ~ j67:1-~ ... 11::~7° 7 /' ,,0)3{tj§~ ... 71<!Jl1f.m1Jffi

~O)~~~U[tJ

c'tlk

tJ:5t!f-c'JDffl ~;h -C~, {) iJi ... 3{tj§0)~~,Um11::0) t~~ ~~~:rL1!l) ;A'}viJi~

m ~;h {)t~~iJi~ ~

'0

~~ G 'iiBi1'J-c'O)3{tj§O)d.tib:JfJ1i~ C -'C O)~,t~ ... ~i§Pjtd.t~~ICJ{$ ~;h-c

~-r ~ 7~{*0)d.tib~'t~~~OO-t {) ~~~1T~

' ...

-'C O)*5~~¥R'6-L -c ~'{) iJi 3 flmO)d.tib:JfJ~O) j!~ ~~ --:) ~'-c 'i+:5ttJ~DJ!iJi1~ G;h -c ~,tJ ~

'0

*{i]f~-c"i ... raJ~Itl::"T';t7J} 7 (MEMRECAM Ci/RX-1) ~{JemL-c~j§O)~TJX ...

~{* ... Jm~~U~ L ... d.tib:JfJ1i~O)j!~~{tt~~ --:) ~'-Ct~§i L t~o ~~-c' 'i71<iBi1'J ~~~fLj[) ;A'

}vit~'i2*) ;A'JviJ)G~~~~~j6A.d:o ~fLj[) ;A'JviJ)GO)*~i§ (3{t{*A 7 J") O)~

~'i ... ) ;A'JvU.ltb~ItiJi1:t=1-~j§O)I.:n~Iti:-I.iV~~{ttT-c'~c. ~ c. C ... 2*) ;A'JvO)

~.~-c' 'i ... ) ;A'JvtB

01-tJlI07f.

-c' 'i~j§O)~{*iJ~~

t

tJ ~'c. C: iJiB~ G iJ) C: tJ ""? t~o