**6.2 Experimental conditions and methods**

By means of a recycling liquid film apparatus (4 cm in each individual airlift pipe diameter, 1 cm in effective height, 12.56 cm2 in cross-sectional area of the airlift part), liquid film aeration tests in a 53 cm deep, 80 L water tank with a surface area of 1510 cm2 are carried out applying an air flow rate of 6 L/min. As a control, under the otherwise identical experimental conditions, the conventional aeration test with air and diffused test with nitrogen are both investigated in this study. Their experimental apparatuses are shown in Fig. 32.

Improvement of Oxygen Transfer Efficiency in

kLab and kLas are calculated based upon Equation (9).

Diffused test with nitrogen

film aeration trial and 0.94 mg/L for the diffused test with nitrogen.

hence deduced as follow,

**6.4 Results and discussion** 

0.00

3.00

DO (mg/L) 6.00

9.00

Diffused Aeration Systems Using Liquid-Film-Forming Apparatus 367

Steady state is reached between the absorption of oxygen through surface transfer and the stripping of oxygen from the water by nitrogen bubbles, *i.e.*, dC/dt = 0. The formula can be

> *L b sat n Ls n ka C C ka C*

DO concentration versus time curves for liquid film and conventional aeration tests as well as diffused test with nitrogen are illustrated in Fig. 33. The ultimate steady-state DO concentrations are 8.39 mg/L for the conventional aeration trial, 8.72 mg/L for the liquid

Liquid film aeration with air Conventional aeration with air

0 3 6 9 12 15 18

Time (min)

Fig. 33. DO concentration vs. time curves for liquid film aeration and conventional aeration

Table 8 presents the calculation data derived from the experimental results shown in Fig. 33. Under the operating conditions of 80 L in water capacity, 53 cm in aeration depth, 1510 cm2 in surface area and 6 L/min in air flow rate, a (20) *Lk a* value of 8.4 hr-1 for the conventional aeration system is calculated. In contrast, the level of (20) *Lk a* for the liquid film aeration system with the relevant LFFA installed simply on the water surface is 9.6 hr-1 (*cf.* Table 8). As compared with these data, the (20) *Lk a* is increased by about 14% as a consequence of the

tests as well as diffused test with nitrogen (Csat. = 8.84 mg/L)

substantial oxygen transfer contribution from the LFFA.

where Cn (mg/L) is the steady-state DO concentration in the diffused tests with nitrogen.

<sup>−</sup> <sup>=</sup> (9)

Fig. 32. Experimental apparatuses of testing liquid film aeration with air, and conventional aerations with air and nitrogen, respectively

#### **6.3 Calculation methods**

The calculation method for total volumetric mass transfer coefficient (kLa) follows the ASCE Standard for Measurement of Oxygen Transfer in Clean Water [18]. Then the obtained kLa is calibrated to a standard reference temperature of 20 °C by using Equation (5) [16, 18]. When a submerged diffuser is operating, there are two main interfaces through which oxygen transfer occurs, *i.e.*, bubble-water interface and air-water interface, as reflected in

Equation (7) [2, 3],

$$\frac{d\mathbb{C}}{dt} = \frac{k\_L a\_b}{h\_d} \int\_z \left(\mathbb{C}\_o^\* - \mathbb{C}\right) dz + k\_L a\_s \left(\mathbb{C}\_{sat} - \mathbb{C}\right) \tag{7}$$

where kLab (l/hr) is the volumetric mass transfer coefficient for bubble surface, kLas (l/hr) is the volumetric mass transfer coefficient for water surface, hd (m) is the depth from diffuser to water surface, z (m) is a variable distance from the diffuser, Csat. (mg/L) is the saturation oxygen concentration in water at atmospheric pressure, C (mg/L) is the actual DO concentration in water body and Co\* (mg/L) is the liquid-phase equilibrium oxygen concentration of a bubble. Co\* is not only a function of temperature and atmospheric pressure, but also hydrostatic pressure and gas-phase oxygen composition. Over depth, the bubble transfer of all gases affects the gas-phase oxygen composition and the equilibrium oxygen concentration.

Wilhelms and Martin developed an approach to split surface and bubble transfer by releasing nitrogen gas from a diffuser rather than air [1]. Herein, it is assumed that no oxygen is initially present in the bubbles, thereby eliminating the estimation of a value for concentration inside the bubbles. Additionally, as a result of utilizing the complete and shallow-depth aeration approach in this set of experiments, the contribution of water depth to DO concentration distribution is thus negligible. That is because the DO concentration is considered to be approximately constant over space in the whole water tank. Under such assumptions, Equation (7) can be rewritten as Equation (8),

$$\frac{d\mathbf{C}}{dt} = k\_L a\_b (\mathbf{0} - \mathbf{C}) + k\_L a\_s (\mathbf{C}\_{sat} - \mathbf{C}) \tag{8}$$

Steady state is reached between the absorption of oxygen through surface transfer and the stripping of oxygen from the water by nitrogen bubbles, *i.e.*, dC/dt = 0. The formula can be hence deduced as follow,

$$\frac{k\_L a\_b}{k\_L a\_s} = \frac{\mathbf{C}\_{sat} - \mathbf{C}\_n}{\mathbf{C}\_n} \tag{9}$$

where Cn (mg/L) is the steady-state DO concentration in the diffused tests with nitrogen. kLab and kLas are calculated based upon Equation (9).

### **6.4 Results and discussion**

366 Mass Transfer - Advanced Aspects

Liquid-film-type aeration experiment Air aeration experiment Nitrogen aeration experiment Fig. 32. Experimental apparatuses of testing liquid film aeration with air, and conventional

Air Air Nitrogen

The calculation method for total volumetric mass transfer coefficient (kLa) follows the ASCE Standard for Measurement of Oxygen Transfer in Clean Water [18]. Then the obtained kLa is calibrated to a standard reference temperature of 20 °C by using Equation (5) [16, 18].

When a submerged diffuser is operating, there are two main interfaces through which oxygen transfer occurs, *i.e.*, bubble-water interface and air-water interface, as reflected in

where kLab (l/hr) is the volumetric mass transfer coefficient for bubble surface, kLas (l/hr) is the volumetric mass transfer coefficient for water surface, hd (m) is the depth from diffuser to water surface, z (m) is a variable distance from the diffuser, Csat. (mg/L) is the saturation oxygen concentration in water at atmospheric pressure, C (mg/L) is the actual DO concentration in water body and Co\* (mg/L) is the liquid-phase equilibrium oxygen concentration of a bubble. Co\* is not only a function of temperature and atmospheric pressure, but also hydrostatic pressure and gas-phase oxygen composition. Over depth, the bubble transfer of all gases affects the gas-phase oxygen composition and the equilibrium

Wilhelms and Martin developed an approach to split surface and bubble transfer by releasing nitrogen gas from a diffuser rather than air [1]. Herein, it is assumed that no oxygen is initially present in the bubbles, thereby eliminating the estimation of a value for concentration inside the bubbles. Additionally, as a result of utilizing the complete and shallow-depth aeration approach in this set of experiments, the contribution of water depth to DO concentration distribution is thus negligible. That is because the DO concentration is considered to be approximately constant over space in the whole water tank. Under such

> (0 ) ( ) *L b L s sat dC ka C ka C C*

( ) ( ) *L b* \* *o L s sat <sup>z</sup> <sup>d</sup> dC k a C C dz k a C C*

= −+ − ∫ (7)

= −+ − (8)

aerations with air and nitrogen, respectively

*dt h*

assumptions, Equation (7) can be rewritten as Equation (8),

*dt*

**6.3 Calculation methods** 

Equation (7) [2, 3],

oxygen concentration.

DO concentration versus time curves for liquid film and conventional aeration tests as well as diffused test with nitrogen are illustrated in Fig. 33. The ultimate steady-state DO concentrations are 8.39 mg/L for the conventional aeration trial, 8.72 mg/L for the liquid film aeration trial and 0.94 mg/L for the diffused test with nitrogen.

Fig. 33. DO concentration vs. time curves for liquid film aeration and conventional aeration tests as well as diffused test with nitrogen (Csat. = 8.84 mg/L)

Table 8 presents the calculation data derived from the experimental results shown in Fig. 33. Under the operating conditions of 80 L in water capacity, 53 cm in aeration depth, 1510 cm2 in surface area and 6 L/min in air flow rate, a (20) *Lk a* value of 8.4 hr-1 for the conventional aeration system is calculated. In contrast, the level of (20) *Lk a* for the liquid film aeration system with the relevant LFFA installed simply on the water surface is 9.6 hr-1 (*cf.* Table 8). As compared with these data, the (20) *Lk a* is increased by about 14% as a consequence of the substantial oxygen transfer contribution from the LFFA.

Improvement of Oxygen Transfer Efficiency in

related equipment and operation.

p.1890-1904 , 2003.

*Fed.*, 61, p.208-220, 1989.

411, 2003.

p.273-281, 2004.

p.407-412, 1972.

*Technol.*, 28, p.163-171, 1993.

**8. References** 

Diffused Aeration Systems Using Liquid-Film-Forming Apparatus 369

By exploring a range of design factors affecting oxygen transfer efficiency of the LFFA, various design parameters of the LFFA are basically determined. The optimal structural parameters of the LFFA are identified to be 4 cm in airlift pipe diameter, ca. 1 cm in effective

By forming a liquid film in the LFFA, the water surface based oxygen transfer efficiency increases by a factor of approximately 2.3 in comparison to the conventional aeration system. The overall oxygen transfer efficiency is raised to 14%. Especially, even at a very shallow aeration depth of 0.5 m, its (20) *EA* still reaches up to 6.64%. These data convincingly suggest that LFFA developed here has an effective oxygen transfer capacity. Meanwhile, as a result of forming a liquid film by LFFA, the effluent water of high DO concentration can be produced even at an aeration depth of about 0.5 m. Therefore, the conventional aeration depth can be reduced down to a very shallow extent by installing LFFA. A direct outcome is that the energy consumption from the air supply devices such as the blower is supposed to be greatly lowered, therefore saving the costs of both aeration-

[1] Wilhelms, S. C. and Martin, S. K., Gas transfer in diffused bubble plumes, In Jenning S.

[2] McWhirter, J. R. and Hutter, J. C., Improved oxygen mass transfer modelling for diffused/subsurface aeration systems. *AIChE J*., 35, p.1527-1534, 1989. [3] DeMoyer, C. D., Schierholz, L. E., Gulliver, J. S. and Wilhelms, S. C., Impact of bubble

[4] Peterson, R. R., Design criteria for high purity oxygen treatment of kraft mill effluent. *J.* 

[5] Nelson, J. K. and Puntenney, J. L., Performance comparison of the air and high purity oxygen activated sludge systems. *J. Water Pollut. Control Fed.*, 55, p.336-340, 1983. [6] Stenstrom, M. K., Kido, W., Shanks, R. F. and Mulkerin, M., Estimating oxygen transfer

[7] Yuan, W., Okrent, D. and Stenstrom, M. K., Model calibration for the high-purity oxygen

[8] Kuo, J. F., Dodd, K. M., Chen, C. L., Horvath, R. W. and Stahl, J. F. Evaluation of tertiary

[9] Tzeng, C. J., Iranpour, R. and Stenstrom, M. K., Modelling and control of oxygen transfer

[10] Jang, A. and Kim, I. S., Effect of high oxygen concentrations on nitrification and

[11] Moore, T. L., Basic criteria and design aspects for deep aeration tanks. *Water Res.*, 6,

sludge plant effluent. *Water Environ. Res.*, 69, p.34-43, 1997.

*search of solutions*. ASCE, New York, p.317-322, 1992.

*Water Pollut. Control Fed.*, 47, p.2317-2329, 1975.

M. and Bhowmilk N. G., Eds. *Hydraulic Engineering*: *saving a threatened resource-in* 

and free surface oxygen transfer on diffused aeration systems. *Water Res*., 37,

capacity of a full-scale pure oxygen activated sludge plant. *J. Water Pollut. Control* 

activated sludge process-algorithm development and evaluation. *Water Sci.* 

filtration and disinfection systems for upgrading high-purity oxygen-activated

in high purity oxygen activated sludge process. *J. Environ. Eng*.-*ASCE*, 129, p.402-

performance of high-purity oxygen A/O bio-film process. *Environ. Eng. Sci.*, 21,

height and 1.1 L/min in air flow rate per unit cross-sectional area of an airlift part.

The *L b k a* and *L s k a* of conventional aeration testings are shown in Table 8 (i.e., *L b k a* = 7.5 hr-1 and *L s k a* = 0.9 hr-1). That is, in a water tank with a water loading of 80 L, aeration depth of 53 cm, and surface area of 1510 cm2, while aerating at an air flow rate of 6 L/min, oxygen transfer amount through water surface has a share of roughly 11% of the overall oxygen transfer quantity.

The oxygen transfer of liquid film aeration involves oxygen transfers through gas bubbles, water surface and liquid film. Because the experimental conditions such as aeration depth, aeration amount and gas bubble diameter are identical between the conventional aeration and liquid film aeration, it is reasonable to consider that gas bubble based oxygen transfer capabilities are same in both cases. For the water surface based oxygen transfer ability, liquid film aeration apparatus only occupies a water surface area of 12.56 cm2 out of the total surface area of 1510 cm2, taking up roughly 0.83% of the whole water surface area. Hence, the coverage area from the liquid film apparatus is completely negligible. However, because the disturbance effect of the liquid film apparatus on the water surface can not be estimated (either positive or negative action), in this discussion, water surface based oxygen transfer capacity is hard to be regarded as same in both cases. Accordingly, the liquid film based oxygen transfer capacity can not be accurately quantified. For the liquid-film aeration experiment, the oxygen transfers through the water surface and liquid film are thus summed up as an overall entity.

Based upon the reason above, the *L b k a* and *L s k a* of a liquid-film aeration system are derived as 7.5 and 2.1 hr-1, respectively. Namely, the oxygen transfer amount via water surface accounts for 22% of overall oxygen transfer capacity.

As shown in Table 8, water surface based oxygen transfer efficiency by means of liquid film aeration apparatus is enhanced to 2.3 times in relation to that by the conventional aeration setup.


*kLat*: total volumetric mass transfer coefficient.

*kLab*: volumetric mass transfer coefficient for bubble surface.

*kLas*: volumetric mass transfer coefficient for water surface.

Table 8. Comparative results for *kLa* between the conventional aeration and liquid-film aeration systems.

### **7. Summary**

Through a series of experiments, it is proven that the oxygen transfer rate of liquid-film aeration system is higher than that of conventional aeration system. Furthermore, the former can transiently produce the water with a high DO concentration. Obviously, this efficient aeration process is considered as a less energy-intensive alternative to current aeration methods.

By exploring a range of design factors affecting oxygen transfer efficiency of the LFFA, various design parameters of the LFFA are basically determined. The optimal structural parameters of the LFFA are identified to be 4 cm in airlift pipe diameter, ca. 1 cm in effective height and 1.1 L/min in air flow rate per unit cross-sectional area of an airlift part.

By forming a liquid film in the LFFA, the water surface based oxygen transfer efficiency increases by a factor of approximately 2.3 in comparison to the conventional aeration system. The overall oxygen transfer efficiency is raised to 14%. Especially, even at a very shallow aeration depth of 0.5 m, its (20) *EA* still reaches up to 6.64%. These data convincingly suggest that LFFA developed here has an effective oxygen transfer capacity.

Meanwhile, as a result of forming a liquid film by LFFA, the effluent water of high DO concentration can be produced even at an aeration depth of about 0.5 m. Therefore, the conventional aeration depth can be reduced down to a very shallow extent by installing LFFA. A direct outcome is that the energy consumption from the air supply devices such as the blower is supposed to be greatly lowered, therefore saving the costs of both aerationrelated equipment and operation.
