Assessment of a New Sludge and Floating-Particle Concentrator Involving Sedimentation and Mechanical Recovery Adapted for Fish-Farm Use

Final Report

Ferme Piscicole des Bobines Inc.

AIMAP 2011-Q02

1. Abstract

A new design for a concentrator for aquaculture sludge was built and tested at the Ferme Piscicole des Bobines, located in southwestern Quebec. This fish farm uses the principle of recirculating rearing water in order to reduce the requirement for make-up water to ensure optimal quality of the rearing water. Recirculating water requires mechanical processing, primarily the use of drum filters to clarify the water by continually recovering fish excrement. Drum filters, however, require large volumes of water to clean the filter membrane (backwash water or diluted sludge). At the Ferme Piscicole des Bobines, all the fish-farm filters generate nearly 400 m³/d, which cannot be held entirely in a settling pond. Indeed, it is not very economical to build a settling pond that would be able to hold the total volume of water for an extended period of time, such as throughout the winter. Untreated and concentrated, this diluted sludge will release large quantities of phosphorus into the effluent.

A concentrator using the principle of sludge sedimentation in two settling cones and equipped with an automatic sludge skimmer for floating materials was therefore built and tested in summer 2012. The sludge concentrated in the two cones was recovered using an electronic system controlling the start sequences (intervals between flushing) and the flushing time for each of the two cones (the time that sludge pumping lasts). This electronic system makes it possible to specify the intervals and duration of skimmer operation, make it possible to remove floating materials accumulating on the surface of the concentrator, namely sludge that was hydrolyzed or carry to the surface by gases formed in the two cones. Excluding costs for the building and labor, the total construction cost of the concentrator for the project was about $35,000.

With a concentrator input sequence based on a surface loading rate of 2.1 to 2.5 m³/h/m² over 55% of the time, the concentrator was able to reduce the volume of diluted sludge coming from the drum filters by 35 to 153 times. At the same time, it was able to concentrate 73% to 80% of total phosphorus (P) and 79% to 87% of suspended solids (SS) entering the concentrator. Analysis revealed that it is important to flush the cones on a regular basis and to adequately adjust flushing duration in order to optimize phosphorus recovery using the smallest volume of water. Observation made it possible to determine that flushing time should be adjusted differently at different times of the day, because larger sludge loads arrive at the concentrator at the start and the end of the day.

2. Introduction and Objective

Fish farms, such as Ferme Piscicole des Bobines, that use a system of artificial ponds with recirculating water have equipment using mechanical filtration and sedimentation to remove fish excrement. This treatment equipment is required in order to maintain a higher quality of rearing water and to reduce effluent phosphorus content. In a conventional recirculating-water system, such as that in this project, recovery mainly involves drum filters and flushing the settling system installed on the drains at the bottom of the rearing tank. The use of drum filters nevertheless results in a large volume of backwash water, which, when added to the flushing of the settling systems, becomes diluted sludge (that is, excrement recovered from many hundreds of cubic meters of water per day with low solids concentrations: <0.1%).

Because it is impossible to hold all of this diluted sludge in closed units such as settling tanks or ponds, work in recent years have focused on reducing sludge volumes and stabilizing the phosphorus in the effluent from these settling units. The work upstream from the settling units has been to reduce sludge volumes in order to improve sludge retention capacity. Given the large volumes of water, the only viable alternatives are mechanical settling and filtration. Treating the diluted sludge with chemicals has been deemed inappropriate for use in fish farms. It would require labor dedicated exclusively to this kind of treatment, which requires a great deal of maintenance and surveillance. In addition, the infrastructure and chemical costs are rather high. The work downstream from the settling units has focused on chemically treating the phosphorus in their effluent. Indeed, given the long holding times of the sludge in the tank, the effluent contains few particles but a significant quantity of dissolved phosphorus. Nevertheless, treating large volumes of water remains a challenge.

Sharrer et al. (2008) have described the main fish-farming methods used for treatment systems upstream from the settling tank: tank and settling cones, concentrators, filters, and geotextile bags. Marcotte (2010) tested a relatively expensive belt filter, using compressed air to concentrate the sludge. The filter used made it possible to reduce the sludge volume by 68 to 96 times while retaining 70% of the phosphorus. Depending on configuration, geotextile bags have a total phosphorus recovery efficiency of less than 40% without the use of a coagulant aid (Ebeling et al., 2005). MAPAQ and MDDEP (2008) tested concentration systems using sedimentation to reduce the volume of excrement to accumulate. These test demonstrated the difficulty of managing the presence of materials floating on the surface of the concentrators as well as the difficulty in effectively recovering concentrated sludge in small volumes of water. Indeed, rapid and significant accumulations of floating materials on the surface of concentrators have been observed and fish farms using drum filters. It appears that a certain amount of sludge hydrolysis by drum filters, combined with breakdown of excrement that is settled to the bottom of the concentrator, can account for this. The difficult management of these floating materials would automatically cancel out the concentrator's phosphorus recovery efficiency. When these floating materials were not recovered, the efficiency of recovering total phosphorus from the concentrators dropped from about 80% to less than 30% (with respect to the phosphorus present in the diluted sludge at the concentrator inlet).

As for treatment systems downstream from the settling units, Ebling et al. (2005 and 2006) tested polymers and coagulants in order to precipitate and stabilize dissolved phosphorus. Marcotte (2007) tested hydrated lime and coagulant (ferrous sulfate) to fix the phosphorus in the effluent of a settling tank. It appears that the use of lime makes it possible to recover more than 95% of the dissolved phosphorus, but the process is rather tedious when large volumes of water and continuous flows are involved. Such continuous treatments require too much effort on the part of the fish farm with respect to the time needed for maintenance and surveillance.

The objective of this project therefore was to build a new type of concentrator and validate its efficiency in reducing the volume of diluted sludge generated by fish production. In other words, this meant determining the optimal concentration ratio of diluted sludge and determining the recovery efficiency of total phosphorus and solids in the concentrated sludge, while automatically managing the recovery of the floating materials that form on the surface of this type of system. Sludge concentration is required in order to improve the environmental performance of fish farms, particularly those using recirculating-water systems.

3. Methodology

3.1. Building the Concentrator at the Fish Farm

Ferme Piscicole des Bobines has an annual production capacity of 210 tons of rainbow trout, and has all of its make-up water supplied by an underground source. The concentrator was installed to handle all of the diluted sludge recovered by the treatment system used at the fish farm. The system comprises five drum filters and a flushing system for the bottom drains in the rearing tanks. For filters are used to filter the water flowing out of the rearing tanks; the fifth is used to filter the farm's effluent. A flushing system makes it possible to recover the excrement that has settled in the tank bottom drains a couple of times a day. All of the filter backwash water is fed by gravity to a sump pit. Since this pit is located underground, two centrifugal pumps operating at 3600 RPM are required to transfer the diluted sludge to the concentrator inlet. Depending on the volume of diluted sludge, the volume impounded in the sump pit, and the rate of flow of the transfer pumps, the pumps operate alternately or continuously according to a predefined sequence: Transfer pump 1 operates for about six minutes and shuts off for about five minutes before the second transfer pump starts and runs for about six minutes The sequence continues with transfer pump 1 starting up again about five minutes after transfer pump 2 shuts down. This sequence a flow of material to the concentrator inlet about 55% of the time.

3.2. Concentrator Construction Details

The concentrator has six main components.

• Concentrator structure: The concentrator walls are made of concrete to ensure they remain in plumb. The surface available sludge sedimentation—that is, the space between the discharge section and the diffusers—is 11.9 m². The concentrator was set up in a closed, isolated building near the settling tank.

• Two settling cones: The cones are inverted pyramid with side slopes of 60°. They are made of fiberglass and installed on the building's concrete floor. The internal surface of the cones was covered with a layer of gel with a view to facilitating the movement of the sludge towards the cone apex.

• Discharge section located in the upstream section of the concentrator: Arranged across the entire width of the concentrator, the discharge section disrupts the input velocity of the diluted sludge, thereby promoting its flow into the cones. It also receives the floating materials collected by the skimmer, which pushes the materials against the flow towards the section. To ensure gravity flow of the floating materials towards the settling tank, the outlet of the piping for flushing the cones is directed upstream of the discharge section. If the floating materials—foamy and sticky by nature—aren't diluted with the concentrated sludge, they would quickly clog the discharge section. In order to ensure the best concentration ratio, the circumference of the discharge section must be waterproofed to prevent water from the concentrator from infiltrating.

• Diffusers located in the downstream section of the concentrator: The first diffuser maintains the floating materials at the surface of the concentrator. The second diffuser maintains the water level and forces the clarified water out by overflowing.

• Automated system for flushing the cones: An electronic control system was used to independently adjust the operating sequences of both pumps used to flush the concentrated sludge from the cones. The system provides for specifying the flush time and wait time. For example, one minute of flushing followed by one hour of wait time results in a 61-minute interval between two successive flushes of one cone. This cycle is continuously repeated throughout the day.

• System for skimming floating materials: A skimmer with a mechanical chain drive was installed to skim the floating materials off the surface of the concentrator's settling zone. As with the system for flushing the cones, the skimmer was controlled by an electronic system to adjust its operating duration and frequency. It was also possible to adjust skimmer speed.

3.3. Sampling Sessions and Concentrator Operation

Instantaneous flow rates at the concentrator inlet were continuously measured with an electronic doppler flowmeter (Greyline PDFM 5.0 ). An option makes it possible to determine the total volume of water flowing by the flowmeter and log the total daily volume of diluted sludge entering the concentrator. Concentrated sludge volumes were determined by multiplying the actual duration of the cone flushing cycles by the flows of the two pumps established beforehand.

The operating time and wait time of the flush pumps were adjusted the day before the sampling session. During preliminary testing, it was observed that the time required to flush cone 2 was generally half that required to flush cone 1. Because of its location, cone 1 received most of the sludge. Independently of the cone flushing time, the skimmer system was adjusted to operate for seven minutes every three hours at a travel speed of 5.6 cm/s. Depending on its starting position, the skimmer made four to five complete sweeps of the entire sedimentation area.

The measurements of concentrator efficiency were performed by taking water or sludge samples at three locations:

• Concentrator input (diluted sludge): A valve (Ø 12 mm) installed on the sludge inlet pipe (Ø 100 mm) provided a means for taking manual samples. Given the water pressure caused by the sump-pit pumps, it was deemed that the sample taken from the valve would be representative of the overall flow through the pipe. When taking a sample, about 1 L of diluted sludge was taken from the valve every 20 seconds throughout the entire pumping cycle of one of the two transfer pumps. Considering that a pumping cycle lasted about six min., about 18 L of sludge was accumulated in a single container before being homogenized. A sample was then taken and sent to an accredited laboratory for analysis.
• Concentrator output (clarified water): The same procedure was followed for concentrator output with a valve installed on the clarified-water pipe. The pipe pressure, however, depended on the water column in the section upstream from the concentrator.
•  Cone flushing (concentrated sludge): When taking a sample, about 0.5 L of concentrated sludge was normally taken from the flush-water outlet pipe every five to ten seconds throughout an entire flush cycle. All of the samples were then homogenized so that a representative sample could be taken. This sample was then sent to an accredited laboratory for analysis.

Sessions 1 and 2.1 involve taking measurements over an entire day by sampling the total volume of concentrated sludge flushed from the cones in addition to the sampling conducted at the concentrator's inlet and outlet. The other sessions involve sampling only at the concentrator's inlet and outlet.

4. Results and Analysis

4.1. Assessment of Construction Costs

Excluding the costs for the building and considering that most of the work was performed by the fish farmer (therefore, no labor costs included in the total), concentrator construction cost about $34,100, broken down as follows:

• About $8,900 for the structure;
• About $5,800 for the cones, discharge section, and diffusers;
• About $11,000 for the automated flushing system, and;
• About $8,400 for the skimmer system.

4.2. Total-Phosphorus and Suspended-Solids Recovery Efficiency

During the measurement sessions, efficiencies were obtained by comparing the P and SS loads measured in the diluted sludge and clarified water. As a result of the analysis of data from sessions 1 and 2.1, it was decided not to loads measured in the concentrated sludge. Given the significant differences observed in the sludge concentration between the start and end of the flushing of the two cones (and the relatively short amount of time taken for flushing), the sampling seems to have resulted in an underestimation of recovery of concentrated sludge.

Indeed, the recovery efficiency values for P and SS was from 73% to 80% and from 79% to 87%, respectively. Sampling session 2 yielded better results. With respect to this last session, reducing only the flush duration in session 3 resulted in a reduction in recovery efficiencies. These efficiencies increased, however, during sessions 4 and 5, either flush frequency or the pumping duration of concentrated sludge. In the case of session 1, the fact that flushing occurred only four times during the day yielded a recovery efficiency lower than that for session 2.

The measurements indicate that it would be better to increase the number of flushes per day and to adjust flush duration. The minimum recommendation would be to flush both cones every hour to limit the time available for phosphorus dissolution. During the pumping phase, the objective was to withdraw all of the concentrated sludge while minimizing its volume. A shorter pumping time would not make it possible to withdraw the total amount of settled sludge in the cones and allows the phosphorus in the residual sludge to go into solution.

In the case of session 2.2, the samples were taken during a 24-hour cycle. An increase in phosphorus load was observed at the concentrator inlet and outlet at the start and finish of the day. The recovery efficiencies varied in the same way during the day. The variations in loads and recovery efficiencies appear to be related to fish metabolism. During the sampling sessions, the fish were fed four times between 7:30 a.m. and 5:30 p.m. According to Eding (2005), fish produce most of their ammonia waste from eight to sixteen hours after feeding. By estimating that the time for producing phosphorus waste and SS would be similar to that for ammonia waste, it would be of interest to consider the possibility of adjusting cone flushing time according to time of day.

4.3. Concentration Ratio

Considering that the sessions during which recovery efficiencies were optimal, the concentrator made it possible to reduce the volume of diluted sludge by 35 to 50 times. That demonstrates that it is possible to achieve the same phosphorus recovery efficiency at different concentration ratios provided that the cones are flushed longer or more often.

4.4. Efficiency of the Skimmer System

During sampling session 2.2, it was observed that the amount of floating materials on the surface of the concentrator was proportional to the SS load in the diluted sludge: maximum values achieved at the start and end of the day. During this session, it was observed that the daily load of total phosphorus in the floating materials represented about 6% of the daily total phosphorus load in the diluted sludge. About 200 L of floating materials, containing about 1000 mg/L of total phosphorus, were recovered by the skimmer system. Each complete skimming cycle of the total surface of the concentrator therefore sent a volume of about 5 L to the discharge section. It was also observed that increasing the interval between cone flushes (sessions 1 and 5) resulted in an increased volume of floating materials. Independently of the sampling sessions, the initial skimmer adjustment (that is, four or five skimming cycles of the surface every three hours) always made it possible to eliminate all of the floating materials.

5. Conclusion

With a concentrator input sequence based on a surface loading rate de 2.1 to 2.5 m³/h/m² over 55% of the time, optimal concentrator operation was able to reduce the volume of diluted sludge while concentrating 76% of total phosphorus and 83% of suspended solid. Analysis revealed that it is important to flush the cones on a regular basis and to adequately adjust flushing duration in order to optimize phosphorus recovery using the smallest volume of water. Flushing the cones once every hour on a continuous basis appears to be the optimal rate. Nevertheless, observation made it possible to determine that flushing time should be adjusted differently at different times of the day, because larger sludge loads arrive at the concentrator at the start and the end of the day. The current system for controlling flushing, however, does not provide for making such an adjustment. In order to increase recovery efficiencies while maintaining the best concentration ratio, the design of the next concentrator could include an optical probe system in order to determine the optimal flushing duration.

Lastly, the project made it possible to develop a new concentrator that will be highly useful in improving the environmental performance of fish farms, primarily those using recirculating-water systems and generating significant volumes of diluted sludge. Using this type of concentrator system now makes it possible to treat and retain a large proportion of the phosphorus recovery from these rearing systems.

6. References

Ebeling J.M., K.L. Rishel, and P.L. Sibrell. 2005. Screening and evaluation of polymers as flocculation aids for the treatment of aquacultural effluents. Aquaculture Engineering, volume 33, pages 235–249.

Ebeling J.M., C.F. Welsh, and K.L. Rishel. 2006. Performance evaluation of an inclined belt filter using coagulation/floculation aids for the removal of suspended solids and phosphorus from microscreen backwash effluent. Aquaculture Engineering, volume 35, pages 61–67.

Eding, E. 2005. Factors affecting biofiltration. Aquaculture Recirculation Technology Workshop, August 2005, Norway.

Marcotte, D. 2010. Évaluation du filtre à courroie de la compagnie Salsnes pour concentrer les boues d'une pisciculture en recirculation. SORDAC. Technical note 2010.2.

MAPAQ et MDDEP. 2008. Internal documents. Essais réalisés à la Ferme Piscicole des Bobines sur l'efficacité de traitement du bassin d'accumulation des boues : essais avec de la chaux, essais avec sulfate ferrique, essais de concentrateurs, caractérisations des rejets. Ministère de l'Agriculture, des Pêcheries et de l'Alimentation du Québec (MAPAQ) and Ministère du Développement Durable, Environnement et Parcs du Québec (MDDEP).