The Progressive Fish-Culturist 52:41-45, 1990

Oxygen Transfer Efficiency of the Aquatector and

Application to Warmwater Fish Culture

Gerald T. Klar1 and Nick C. Parker2

U.S.Fish and Wildlife Service

Southeastern Fish Cultural Laboratory

Route 3. Box 86

Marion, Alabama 36756, USA

Abstract.—Oxygen transfer rates and efficiencies were determined for a model 32/110 Aquatector—a commercially available device that mixes water and oxygen under pressure, shears the oxygen into microbubbles, and discharges the mixture—at three pressures and six oxygen flows. Standard oxygen transfer rates peaked at 2.87 g/min with 82.8% efficiency at 5.80 kPa, at 1.60 g/min with 33.6% efficiency at 2.90 kPa, and at 1.22 g/min with 21% efficiency at 1.45kPa. Own transfer efficiency was correlated positively with pressure and negatively with oxygen flow. The Aquatector was capable of saturating deoxygenated water at the rate of about 340 L/min at 21 C and 5.80 kPa with 82.8% efficiency of oxygen used. Although the total mass transfer of oxygen from the 32/110 Aquatector was too low to be effective in providing emergency aeration in large volumes of water, such as those in warmwater ponds, larger models (40/160 or 50/225) might be practical.

Inadequate dissolved oxygen often limits production in aquaculture. Dissolved oxygen may be provided by a continuous flow of fresh water, photosynthesis of plants, mechanical aeration, or direct injection of gaseous oxygen into water. In static warmwater culture ponds, oxygen is provided primarily by photosynthesis. Under certain conditions, however, the demand for oxygen exceeds the supply and fish loss may ensue unless emergency aeration is provided

Oxygen levels can be increased by airlift pumps (Parker and Suttle 1987) or surface agitators (Boyd and Ahmad 1987) that increase the exposure of water to atmospheric oxygen. Methods used to inject pure gaseous oxygen into water include diffusers, packed columns, and U-tube systems (Speece 1981; Watten and Beck 1985). Systems that use pure oxygen have been cost effective when 50% or more of the oxygen has been absorbed by the water (H. Westers, Michigan Department of Natural Resources, personal communication).

 

1 Present address: U.S. Fish and Wildlife Service, Marquette Biological Station, 446 E Crescent Street, Marquette, Michigan 49855, USA

2 Present address U. S Fish and Wildlife Service, Texas Cooperative Fish and Wildlife Research

Unit, Texas Tech University, Lubbock, Texas 79409-2125, USA

 

A new device with the trade name Aquatector (Zeigler Brothers, Inc., Gardners, Pennsylvania) has recently been made available to aquaculturists. It mixes water and oxygen under pressure, shears the oxygen into microbubbles, and discharges the mixture. Oxygen transfer efficiencies up to 93% have been reported when an Aquatector was operated with a continuous supply of fresh water (Schutte 1988). No tests of an Aquatector have been reported in which unoxygenated water was used for the measurements of oxygen transfer, as recommended by the American Society of Civil Engineers (ASCE 1984).

The objectives of this study were to use the recommended methods of ASCE (1984) to determine oxygen transfer rates and efficiencies of the Aquatector and to determine its application to warmwater fish culture.

Methods

An Aquatector (model 32/110) rated to deliver oxygen at 3.5 L/min with a water flow up to 50 L/min was placed on a wooden frame over a fiberglass tank 1.9 m in diameter and 0.9 m deep (Figure 1). A 559-W submersible well pump was placed inside the tank and connected to the Aquatector with polyvinyl chloride pipe. Water pressure to the Aquatector was regulated by adjusting a valve that bypassed water from the pump and released it laterally beneath the surface of the water in the tank. The Aquatector was positioned so that its discharge could be directed inside or outside the tank. The discharge of a 246-W submersible sump pump inside the tank was directed laterally to provide circulation. Cooling coils connected to two 246-W compressors were placed in the tank to provide a stable temperature and to offset heat produced by the water pumps.

Two YSI model 57 oxygen meters (Yellow Springs Instrument Company, Yellow Springs, Ohio) were used to measure dissolved oxygen. One meter had an expanded scale (4 ) that measured dissolved oxygen concentrations within the range of 20-40 mg/L. The second meter measured oxygen in the range of 0-20 mg/L. Both meters were connected to a two-channel Omniscribe chart recorder (Houston Instruments, Austin, Texas). Winkler titrations, modified for saturated and supersaturated dissolved oxygen (APHA et al. 1985), were used to calibrate the meters. Before each test, both meters were calibrated to a water sample at oxygen saturation, and the linearity was verified by Winkler titration to 10% saturation; linearity to supersaturation was verified during the test. After each test, the calibration to saturated oxygen was again verified. The meters were within 5% of the Winkler titration values for all tests.

fig1.gif (10688 bytes)

FIGURE 1.—Components of the system used to evaluate the Aquatector (A) oxygen bottle with regulator, (B) adjustable gas flowmeter, (C) oxygen line, (D) water displacement column used to calibrate gas flowmeter, (E) overflow pipe, (F) check valve, (G) Aquatector, (H) needle valve, (I) 19-L bucket, (J) 1,710-L tank, (K) dissolved oxygen meter and probe, (L) chart recorder, (M) water supply pump for Aquatector, (N) water lime, (O) water supply pump for gasometer, (P) gasometer, (Q) microbore mercury manometer.

 

Oxygen flow to the Aquatector was regulated with a precision flowmeter (model F. 6360, Gilmont Instruments, Inc. Great Neck, New York). A gauge was used to monitor the gas pressure at the outflow of the flowmeter. The oxygen flowmeters were calibrated by water displacement per unit of time at each pressure and flow setting tested (Table 1).

We began a typical test sequence by pumping water into the Aquatector to flush all air from the system. The oxygen flow was then started and pressure adjusted to approximately that required for the three tests—1.45, 2.90, and 5.80 kPa.3 After the oxygen flow had stabilized, we adjusted the water outflow to about 28 L/min; the oxygen pressure and flow were then set for each test. These adjustments were accomplished by manipulating water flow into and out of the Aquatector. During these adjustments, the water outflow from the Aquatector was diverted outside the test tank and fresh water was added to the test tank to maintain a static water level. While the Aquatector was equilibrating, the oxygen meters, with outputs connected to the chart recorder, were calibrated and probes were positioned in the test tank After equilibration, the Aquatector discharge was directed into the test tank, the flow of fresh water to the tank was stopped, and water volume was adjusted to 1,550 L. Styrofoam chips (packing material) were layered over the water surface to a depth of 8 cm to essentially eliminate the introduction of atmospheric oxygen resulting from surface turbulence. Reagent-grade cobalt chloride was added to the test tank to yield a concentration of 0.50 mg cobalt/L. After 10 min. sufficient technical grade sodium sulfite was added to reduce the oxygen content to zero for at least 2 min. Reoxygenation of the test tank was then recorded until readings were stable.

_________

3 One kPa = 6.895 Ib/in2.

table1.gif (2353 bytes)

TABLE 1.—Oxygen flow rates (g/min) tested in a model 32/110 Aquatector at three pressures in six trials.

Oxygen transfer curves were developed by using the nonlinear regression model given by ASCE (1984). Water temperatures were 21 1C at the beginning of each test and increased no more than 3C during the test. Ambient barometric pressure was 760 3 mm Hg for all tests. Oxygen transfer rates and efficiencies were calculated from the regression model at ambient conditions for each test.

Results

The rate of aeration of completely deoxygenated water by an Aquatector increased as oxygen pressure increased from 1.45 to 5.80 kPa and oxygen flow increased to about 3.5 g/min (Figures 2-4). Standard oxygen transfer rates peaked at 2.87 g/min with 82.8% efficiency (Figures 5, 6) when oxygen flow was 3.47 g/min at 5.80 kPa and at 1.60 g/min with 33.6% efficiency when oxygen flow was 4.78 g/min at 2.90 kPa. Standard oxygen transfer rates increased to 1.22 g/min with 21% efficiency when oxygen flow was 5.82 g/min at 1.45 kPa. Oxygen transfer efficiency was positively correlated with pressure at oxygen flow rates greater than 1 g/min; transfer efficiency decreased when oxygen flow increased and pressure remained stable (Figure 6).

fig2.gif (2846 bytes)

FIGURE 2.—Reoxygenation of water by a model 32/ 110 Aquatector operated at 1.45 kPa pressure (DO = dissolved oxygen). Solid lines are regression lines plotted on every fourth data point recorded. Oxygen flow rates (g/min) were as follows: (A) 0.429, (B) 1.01, (C) 1.98, (D) 3.48. Curves for oxygen flow rates of 4.73 and 5.82 g/min were very close to curves C and D.

fig3.gif (2947 bytes)

FIGURE 3.—Reoxygenation of water by a model 32/110 Aquatector operated at 2.90 kPa pressure (DO = dissolved oxygen). Solid lines are regression lines plotted on every fourth data point recorded. Oxygen flow rates (g/min) were as follows: (A) 0.426, (B) 1.03, (C) 1.98, (D) 3.52. Curves for oxygen flow rates of 4.78 and 6.08 g/min were very close to curves C and D.

fig4.gif (3070 bytes)

FIGURE 4.—Reoxygenation of water by a model 32/110 Aquatector operated at 5.80 kPa pressure (DO = dissolved oxygen). Solid lines are regression lines plotted on every fourth data point recorded. Oxygen flow rates (g/min) were as follows: (A) 0.427, (B) 1.02, (C) 1.95, (D) 3.47. Curves for oxygen flow rates of 4.76 and 5.94 g/min were close to curves C and D.

fig5.gif (2865 bytes)

FIGURE 5.—Comparison of standard oxygen transfer rate for an Aquatector model 32/110 operated at three pressures (kPa): (A) 1.45, (B) 2.90, (C) 5.80.

Discussion

The Aquatector tested appears to be capable of adding oxygen to small quantities of water with high efficiency. Model 32/110 saturated about 340 L of deoxygenated water per minute at 21C and 5.80 kPa with 82.8% efficiency of oxygen used. This level of oxygenation would be adequate for a limited water flow through a recirculation system or in a small flow-through system in which oxygen requirements are fairly stable. Aquatectors may also be effective in supplementing downstream oxygen requirements in flow-through culture of salmonids.

Channel catfish (Ictalurus punctatus) are commonly reared in 2-8-hectare ponds, and ponds of other sizes may be used to culture other warmwater fishes. Emergency aeration mechanisms for commercial-sized warmwater culture ponds must be capable of quickly adding oxygen to large volumes of water. If the concentration of oxygen in intensively stocked channel catfish ponds drops below 2 mg/L, the level must be increased to above 3 mg/L within 30 min to prevent stress and loss of fish. About 72 model 32/110 Aquatectors would be required to add 2 mg oxygen/L to 1 hectare of water I m deep in 30 min at 5.80 kPa and 82.8% efficiency. Larger models (40/160 and 50/225) that have been used in commercial fish farms in Chile (R. Piper, Bozeman, Montana, personal communication) might effectively provide emergency oxygenation in large volumes of water. (The model 50/225 Aquatector is 4 times larger than model 32/110.)

Several of the calculated oxygen transfer efficiencies at low oxygen flows were slightly above 100% (Figure 6). The one exception was at 5.80 kPa, when calculated efficiencies increased from below 100% to above 100% and then declined as oxygen flow increased. Small errors in oxygen flow measurements and oxygen concentration may account for the efficiencies calculated to be above 100%. The manufacturer recommends that the oxygen and water interface in the sight chamber of the Aquatector be maintained at an intermediate level for best operation. This level appears to be at oxygen flow rates that produce about 80% transfer efficiencies at the three pressures tested. Higher oxygen flows were required to record this level at 5.80 kPa than at 2.90 or 1.45 kPa. Operating the Aquatector at high pressure and greatly reduced oxygen flow increased oxygen transfer efficiency.

fig6.gif (2899 bytes)

FIGURE 6.—Oxygen transfer efficiency for a model 32/110 Aquatector operated at three pressures (kPa): (A) 1.45, (B) 2.90, (C) 5.80.

 

Conclusion

The model 32/110 Aquatector appears to be best suited to provide continuous oxygenation in warmwater aquaculture systems with small volumes of water. This model is not suitable as an emergency aeration device for commercial rearing ponds.

Acknowledgments

For reviewing the manuscript, we thank W. Krise, U.S. Fish and Wildlife Service, Wellsboro, Pennsylvania, and H. Westers, Department of Natural Resources, Lansing, Michigan. We thank T. Zeigler, Zeigler Brothers, Inc., Gardners, Pennsylvania, for providing the Aquatector used in this study.

References

  • APHA (American Public Health Association), American Water Works Association, and Water Pollution Control Federation. 1985. Standard methods for the examination of water and wastewater, 16th edition. American Public Health Association, Washington, D.C.

    ASCE (American Society of Civil Engineers). 1984. A standard for the measurement of oxygen transfer in clear water. ASCE, New York.

Boyd,C. E., and T. Ahmad. 1987. Evaluation of aerators for channel catfish farming. Alabama

Agricultural Experiment Station, Bulletin 584, Auburn University, Auburn, Alabama.

Parker, N. C., and M. A. Suttle. 1987. Design of airlift pumps for water circulation and aeration in

aquaculture. Aquacultural Engineering 6:97-110.

Schutte, A. R. 1988. Evaluation of the Aquatector : an oxygenation system for intensive fish

culture. Progressive Fish-Culturist 50:243-245.

Speece, R. E. 1981. Management of dissolved oxygen and nitrogen in fish hatchery waters. Pages

53-62 in L. J. Allen and E. C. Kinney, editors. Proceedings of the big-engineering symposium

for fish culture. American Fisheries Society, Fish Culture Section, Bethesda, Maryland.

  • Watten, B. J., and L. T. Beck. 1985. Modeling gas transfer in a U-tube oxygen adsorption system: effects of off-gas recycling. Aquacultural Engineering 4:271-297.