THE LONG PATH, PLUG FLOW, SUBMERGED
BIOLOGICAL FILTER
Introduction
Operators of recirculating aquaculture
systems are blessed with a wide variety of choices when it comes to biological
filters. Most of the filters used are
either well mixed reactors with medium to long retention times or short path
plug flow reactors with relatively short retention times. These systems are sufficiently efficient and
effective when used for raising crops like tilapia that are tolerant of poor
water quality. The problems start when
farmers try to raise crops that are not tolerant of poor water quality. The
biofilters commonly used today are very inefficient when very low levels of
ammonia are required.
The tough part about maintaining the
water quality necessary to raise delicate species is doing it while feeding the
fish at the levels necessary to produce a crop in an economic time frame. It is not that difficult to maintain water
quality while feeding at .5% of body weight in a system stocked at .1 lb of
fish/gal of water. It is entirely
different to maintain water quality while feeding at 3% of body weight in a
system stocked at .75 lb/gal. It is the
difference between a hobby aquarium and a profitable recirculating aquaculture
system.
For the foreseeable future, the most
intelligent solution for biofiltration and suspended solids control in
recirculating systems will be to employ several different types of water
quality control devices simultaneously.
Each device should have one primary function. This has numerous
advantages.
1. Each individual device can be simpler.
2. Simpler devices tend to be more reliable.
3. In the event of a malfunction or failure
of a single device, the overall effect on the system will not be as great. In some cases, other components of the system
can compensate for the loss of one.
4. Downtime for maintenance of one component
in a multicomponent system has less of an effect than losing the "do it
all” type biofilter/solids filter.
5. If a multi component strategy is adopted,
it is easier to add more components as the need arises.
6. Each type of biofilter or solids control
device has its own characteristic advantages and weaknesses. By using several different devices
simultaneously these devices can compliment each other and do a better overall
job of maintaining water quality.
7. Using numerous devices increases the
flexibility of the overall system and enables the farmer to change crops as the
market dictates the need.
Types of reactors and their characteristics
There are two basic types of
biological and chemical reactors: continuously stirred tank reactors (CSTR) and
plug flow reactors. Continuously stirred
tank reactors are also known as well mixed reactors. CSTR or well mixed
reactors are easily visualized as vessels or tanks that are stirred to achieve
uniformity throughout the tank.
A very important characteristic of the
CSTR is that the concentration of the reactants in the outlet is equal to the concentration
of the reactants in the vessel regardless of the concentration of the reactants
in the inlet.
Plug flow reactors can be modeled as a
pipe where the reactants move as a plug along the pipe. The concentrations of the reactants will vary
along the pipe and there is no mixing between the beginning and the end of the
system.
One of the obstacles facing
recirculating systems trying to culture species with high water quality needs
is the current reliance on CSTR bioreactors or short path plug flow reactors.
Here are some examples of these types of systems.
Moving
Bed Biofilter (MBB) - This type of biofilter is a good example of a CSTR. It adds oxygen to the water during normal
operation and it has a very low operating head.
A number of small MBB's arranged in series could approximate a long path
plug flow reactor.
Fluidized bed
sand filters (FBSF) – These filters are half way between CSTR and plug flow
biofilters. If a system were designed with numerous smaller reactors in series,
the system may approach the operating characteristics of a plug flow
reactor. However, the high pressure drop
of FBSF's make this possibility economically impractical.
Bead Filters –
These are primarily short path, plug flow reactors. Unfortunately they suffer from high pressure
drop and cyclical performance as they load up with solids and require
backwashing. The major problem with bead
filters is the conflicting and mutually exclusive tasks of collecting solids
and removing ammonia. The collection of
solids encourages the growth of heterotrophic bacteria at the expense of
nitrifying bacteria.
Trickling
Filters - These
are very good plug flow reactors but the path is short. The short path drawback can be overcome by
using several trickling filters in series but the energy required for all the
pumping would make it uneconomical.
Trickling filters are the best all around biofilters due to their
ability to do gas exchange and biofiltration.
Their only real drawback is the pump head required to lift the water up
to the top of the filter.
Rotating
Biological Contactors – RBC's have the potential to approximate a long path,
plug flow reactor if multiple units are arranged in series. Since RBC's have very low pump head
requirements it is very feasible to use them as a LPPFR.
Submerged Bed
Packed Filters – These types of filters operate best as long path, plug flow
reactors. They have very low pump head
requirements and operate hydraulically better with a long flow path and long
retention times. Traditionally,
submerged filters have been upflow or downflow.
However, horizontal flow is the best orientation for a long path, plug
flow reactor.
Comparison of CSTR vs. Plug Flow Reactors
If the species being cultured can
tolerate fairly high concentrations of ammonia, a CSTR can be an excellent
choice for a biofilter. Above 2.5 ppm
TAN (total ammonia nitrogen), ammonia removal rates are constant and
independent of ammonia concentration. Therefore, there is no difference in
overall efficiency between well mixed reactors and plug flow reactors if the
desired outlet concentration of the biofilter is 2.5
ppm TAN or greater.
Below 2.5 ppm TAN, the ammonia removal
rate starts to become dependent upon the ammonia concentration. The rate at
which biofilters remove ammonia is proportional to the square root of the
concentration of ammonia. This is commonly known as a 1/2 order reaction
rate. The following graph shows a
typical ammonia removal curve.
In a well stirred reactor, the
influent water is quickly diluted so that the ammonia concentration is equal to
the outlet concentration. Thus, if the
inlet concentration is 1.0 ppm TAN and the desired outlet concentration is 0.1
ppm TAN, the bioreactor must be sized based on the very low removal efficiency
at 0.1 ppm.
In contrast to a CSTR, a plug flow
reactor starts off with a removal efficiency that corresponds to the inlet
concentration. The efficiency decreases
through the reactor as ammonia is removed and the concentration drops. Only at the very end of the flow path will
the efficiency of ammonia removal be down to the removal efficiency of the
whole well mixed reactor.
One way to look at the advantage of a LPPF
biofilter versus a CSTR is to look at the amount of surface area required to
accomplish a given ammonia removal rate.
The following graph shows the increasing advantage of the LPPF as the
fraction of ammonia removed increases.
If the cultured species is sensitive
to ammonia and low concentrations are desired, the plug flow reactor can be
significantly more efficient than a CSTR .
Another advantage of a LPPFR, is the
establishment of sequential zones with different bacteria predominating. The inlet section of the biofilter will tend
to have heterotrophic bacteria. After
the carbonaceous oxygen demand (COD) has been absorbed, the nitrosomonas
bacteria will be able to establish themselves and oxidize the ammonia to nitrite. The production of nitrite will allow the nitrospira to establish themselves as a third zone. These zones will not be sharply delineated
but will gradually progress from one to another.
Physical Description
In the past, a common way to construct
submerged filters was to build large flat beds of media with water flow either
upflow or downflow. This is the worst
way to build a submerged filter. It
virtually guarantees plugging, channeling, short circuiting of water and
overall poor performance.
What does a long path, plug flow
bioreactor look like? At the present
time, the most practical method of constructing a long path, plug flow system
is to use a horizontal flow, submerged filter.
The construction of tall, narrow, vertical columns is an expensive way
to build biofilters. Horizontal raceways
are much easier to build and less expensive.
There are a number of possible configurations for a submerged filter
operated as a long path, plug flow reactor.
The simplest is a long raceway.
The
next obvious possibility is a folded raceway.
A
tank can be baffled any number of times to achieve the desired flow path
Although
it is not optimum in terms of tank utilization, a folded vertical system is
also an option.
All
of the above assume a rectangular tank.
Circular tanks can also be used.
By using concentric tanks, an annulus can be created that is an
excellent long path plug flow reactor.
Starting with a 12 ft. diameter tank, the length of the filter would be
almost 38 ft long. For a 20 ft. diameter
tank the length would be almost 63 ft.
Pressure
drop through a LPPFR using structured media is very low. The current trend
toward low head systems is easily satisfied by a LPPFR. The graph below shows head loss at different
superficial face velocities. These
results are for clean media. Biogrowth on the surface of the media will increase the
resistance to flow but the total pressure drop will still be very low. Based on observations and measurements, we
try to design systems with velocities between 4 and 10 ft/min (1.2 – 3 m/min)
Besides the energy savings, another
benefit to a low head system is the ability to accommodate a large flow of
water. A low turnover time for the
culture tank is an important factor when trying to keep ammonia concentrations
low.
Low head systems often use air lift
pumps to move water. One of the additional possibilities to increase the flow
rate through the system and accommodate higher head losses is to use multiple
air lifts at different stages of the reactor.
The additional aeration is another benefit. Since nitrification rates are also dependent
on oxygen concentrations, intermediate air lifts can add to the overall
efficiency and insure high oxygen concentrations when the water is returned to
the culture tank.
The key design feature of a LPPFR that
insures plug flow is a sufficiently high velocity. This creates turbulent flow and prevents back
mixing. Creating a long thin channel for
water flow has a number of practical benefits.
1.
High water velocity insures plug
flow.
2.
Ammonia removal rates are
proportional to Reynolds number so higher water velocities mean higher ammonia
removal rates.
3.
Plug flow means that retention times are
uniform and all of the water is treated for the same amount of time.
4.
There are no complicated measures
necessary to achieve an even water distribution.
5.
The design is both flexible and
simple.
Media Selection
One of the advantages and
disadvantages of a LPPFR using structured media is the ability of the biofilter
to remove very small particles. Biofilms
are very "sticky" and tend to trap small particles. For this reason, correct selection of the
media is very critical. Ideally, the
media will be continue to work regardless of the solids collected. A high void fraction and large free passage
diameter is essential. For this reason,
gravel, fiber pads and ribbon bundles are not recommended. Random dumped plastic media can be used but
the high cost and difficultly of handling make it a poor choice. The best type
of media is a structured media with sheets or plates that are parallel to the
water flow direction. This type of media
is critical to avoid plugging and keep the head loss low. A typical example of a cross corrugated structured
media is shown here.
When sizing a biofilter, it is always advantageous
to use a media with the most surface area per unit volume provided that it will
not plug. There is no rule however that
requires us to only use one type of media.
In order minimize the size of the biofilter and the maintenance
requirements, it is recommended that the packing density vary through the
biofilter. The inlet section should have
the most open or least dense media and following sections can use progressively
denser media.
Summary
The Long Path Plug Flow Bioreactor
using structured media has a number of significant advantages.
1. Low pump head loss minimizes energy
costs.
2. Low pumping costs allow high volume flows
and quick tank turnover times.
3. A long path plug flow reactor requires
less surface area than a CSTR at low ammonia concentration levels.
4. Plug flow insures consistent treatment of
all of the water passing though the biofilter.
5. Easily removed packing allows convenient
observation of operating conditions within the biofilter.
6. Flexible designs can accommodate various
farm layouts.
7. The LPPFB will collect very fine solids
that are normally difficult to remove.
8. Ability to remove ammonia to very low
levels permits flexibility of changing crops.
References
Greiner, A. D., Timmons, M. B., 1998.
Evaluation of the nitrification rates of microbead
and trickling filters in an intensive recirculating tilapia production
facility. Aquacultural Engineering pp 189 - 200
Kamstra,
A., Van der Heul, J.W., Nijhof,
M., 1998.
Performance and optimization of trickling filters on eel farms. Aquacultural Engineering pp 175-192
Saucier, B., Chen, S., Zhu, S.,
“Nitrification Potential and Oxygen Limitation in Biofilters” presented at the Third International
Conference on Recirculating Aquaculture July 2000.
Timmons, M.B., Losordo,
T. M., 1994. Aquaculture Water reuse Systems: Engineering Design and Management Elsevier
Science B.V.
Zhu and Chen “An experimental study on
nitrification biofilms performances using a series reactor system” Aquacultural
Engineering 1999 Vol 23, p. 245 – 259.
©2002-2003
by L. S. Enterprises. All rights
reserved. No part of this publication may be reproduced or transmitted in any
form or by any means electronic or mechanical, including photocopy, recording,
or any information storage and retrieval system, without permission in writing
from the publisher.
Published
by L. S. Enterprises
PO Box 13925
Gainesville,
FL 32604 USA
Author:
Matt Smith
Office 1-352-379-5626
Mobile
1-239-851-1175
Fax
1-866-706-1775
Email:
mattsmith@biofilters.com
Rev. 8/07/2013