by Stephan M. Tanner

Introduction

To keep fish healthy, two parameters are absolutely essential, water quality and nutrition. The two parameters are also closely connected. Adding food to the aquarium provides the necessary energy for the fish to live and grow but it adds “waste” to the tank (and with some feeds more than with others, but that is a separate topic). The fish digest the proteins, nucleic acids, carbohydrates, and fats and excrete carbon dioxide and ammonia via the gills, and defecate undigested material that can contain large amounts of energy, sometimes 50% of that of the original food. Fecal matter therefore contains large amount of nitrogen and phosphorous compounds. While the carbon dioxide mostly diffuses into the air, ammonia and other nitrogen compounds and phosphate accumulate in a closed system like a fish tank. Ammonia, phosphate, and carbon dioxide are vital plant nutrients if the tank contains live plants, however, the amounts of nitrogen and phosphate usually exceed the need and their levels increase. High nitrogen and phosphate levels not only promote unwanted algae growth, they rapidly become toxic to the fish, invertebrates, and even the plants if they exceed certain levels. At this point, there are two solutions, replace the water and/or filter it. Some systems replace the water continuously and automatically (a flow-through system), so filtration may be unnecessary. However, water has become expensive and sometimes limited. Therefore, in most aquaria a filter is installed to purify the water.

I would like to emphasize that I am an advocate of regular water changes no matter how good your filtration is, but a filter reduces the need for continuous water changes and preserves and guarantees the appropriate water quality between changes. There are chemical filtration media such as activated carbon, resins, and zeolithes that absorb nitrate and phosphate but they quickly burn a hole in your wallet since they have to be replaced regularly when they are saturated, and they are not cheap. Chemical additives that bind phosphate, ammonia, and nitrate don’t solve the problem because they only temporarily detoxify the compounds.

Mechanical filtration is not the primary purpose for an aquarium filter, so the term "filter" is not the best description. The primary purpose is actually biodegradation, which is performed by bacteria and other microorganisms that form a community. They live in the filter and turn it into a bioreactor. The filter is there mainly to provide surface area for these microbes, not to mechanically remove debris from the water. That means the best filters will establish good living conditions for these organisms.

Nitrification

Nitrification is the biological conversion of ammonia (NH₄⁺) into nitrite (NO₂^-) and then nitrate (NO₃^-). Ammonia is the most toxic for fish, nitrite is less toxic, and nitrate is the least toxic. This process requires microbes that live in the presence of oxygen (aerobes). A good biological filter containing these microbes is essential to removing ammonia and nitrite from the aquarium. Since nitrate is the final step in nitrification it can build up in an aquarium and levels over 30 ppm can inhibit growth and cause stress in some aquatic species. There are three ways to get rid of it. The first is to dilute it by performing water changes. The second is to have live plants in the tank since plants will utilize the nitrate. The third is to have microbes (nitrifiers) convert the nitrate to nitrogen gas (N₂) in a process called denitrification. The nitrogen gas then dissipates into the air. The process of denitrification requires microbes that live in the absence of oxygen (anaerobes) as will be discussed further below.

As mentioned above, food is eventually turned into ammonia (NH₄⁺) by metabolic processes in the fish. It was long thought that ammonia-oxidizing bacteria (AOB) called Nitrosomonas nitrify ammonia into nitrite and in a second step Nitrobacter oxidize nitrite into the less toxic nitrate. However, you may be surprised to learn that recent research published in the journal PLoS One in August 2011 indicates that ammonia-oxidizing archaea (AOA) are actually the dominant ammonia-oxidizing microorganisms in freshwater aquarium biofilters (Sauder et al 2011). Archaea used to be called Archaebacteria and grouped together with Eubacteria as Bacteria. Now, Archaea are considered a third domain of life apart from Bacteria and Eukaryota. AOA outnumber AOB in many terrestrial and aquatic environments and Sauder and colleagues (2011) provide the first evidence for the important role of AOA in freshwater aquarium filtration. It appears that AOA adapt to the niche of relatively low ammonia conditions found in freshwater aquaria, while AOB are more common in wastewater or sewage treatment plants with very high ammonia concentrations. It is important to note that most commercially available bacteria starter cultures for aquaria are composed of AOB and not AOA. That means if you want to start a new aquarium, it might be better to squeeze the filter sludge from a established aquarium into the new one to jump start the necessary AOA culture. However, the AOB may serve as a stop-gap until the AOA get established. The exact composition of the filter sludge in well-established filters still remains to be studied in detail and we are just beginning to understand how biofiltration actually works. The reason for that is quite simple: so far technology has not permitted the cultivation of many types of microorganisms in isolation (the different types depend on each other to survive), which makes their individual characterization more difficult. Recent improvements in DNA sequencing technology now permit the characterization of entire communities in one shot, both qualitatively (what types are living there) and quantitatively (how many of each type). This new technology is still quite expensive and only gradually some researchers are starting to focus on the aspects that pertain to the freshwater aquarium hobby.

Denitrification

Denitrification is the process in which nitrate is reduced to nitrogen gas. This is performed by microbes that use nitrate in place of oxygen for respiration (denitrifiers). These microbes are called “anaerobic” because they can exist in the absence of oxygen. Some are even killed by oxygen (strict anaerobes). Denitrification can only occur in the absence of oxygen so it was originally thought that this process would not occur in an aquarium. It was thought that since oxygenated water was pulled through the filter medium (undergravel filters, canister filters, sponge filters, etc.), that there was nowhere for anaerobic microbes to live. However, it is now known that microbes grow in communities called biofilms. The inner layers of a biofilm actually lack oxygen because the outer layers of the biofilm use it to respire. Therefore, the inner layers lack oxygen and the microbes there will perform denitrification. The realization of the biofilm concept suggests that even obligate anaerobes exist in the aquarium and the filter media. It is interesting to note that when undergravel filters were very popular, these hobbyists had good success with their fish tanks. The presence of a relatively thick layer of substrate probably enhances the formation of oxygen-depleted pockets that foster the growth of anaerobes. This is another area we have not fully understood and much research remains to be done.

Practical considerations

From the background above we can conclude that the most important component of a filter is the surface for microbes to live on. The question is what and how much. In different types of aquaria (with and without plants, many or few fish, with or without substrate, etc.), different amounts of AOA grow depending on the availability of ammonia. In natural environments, the number of AOA is usually at the upper limit and can barely survive on the available resources, while in the aquarium the levels can vary quickly when food is added to the tank. So the main purpose of the filter is to provide ample microbial biofilms containing both aerobic and anaerobic microbes to consume all of the waste quickly when it appears. Thus, the more surface area the better. However, surface area alone is of no use if the aquarium water (and the waste within) does not contact the microbes. Therefore, only surface area that has a good flow of water going past is going to be useful. You can accomplish this by choosing a material with many holes and capillary passages and a large inflow area.

The second important consideration is ease of maintenance. That depends on two factors, how long does the filter media last between rinsing (=maintenance interval) and the time until it needs to be replaced (=lifespan), both of which affect life cycle cost. Obviously, the longer the better in both cases. To compare various filter media I have listed alphabetically the most commonly sold media on the market today.

Selection of popular filter materials

• Bio balls: easy to rinse, unlimited life, expensive, very low surface area, sludge remains in the filter when lifted out for rinsing.
• Ceramic rings: easy to rinse, unlimited life, heavy, quite expensive, limited surface area, sludge remains in the filter when lifted out for rinsing.
• Filter carbon: chemical filter media that absorbs many impurities, expensive, very short useful life (~2 weeks) and high life cycle cost because the useful life span is very short, non-selective removal of many nutrients, not a long-term biofilter media.
• Filter resins: come in various types for cation or anion exchange or both (amphoteric exchangers), very expensive, needs to be recharged or replaced frequently, not a long-term biofilter media.
• Gravel: cheap, heavy, hard to rinse, difficult to remove from the filter for rinsing, moderate surface area.
• Lava rock: cheap, moderate surface area, difficult to rinse, messy, fine particles wear out powerhead rotors, difficult to remove from filter for rinsing.
• Polyester filter floss and pads: cheap, can be stuffed in almost any filter, difficult to rinse, clogs easily, high life cycle cost because the life span is very short. Not a long-term biofilter media.
• Poret® filter foam: chemically stable, lightweight, long useful life (5-10 years) with the lowest life-cycle cost of any filter media, easy to rinse, very high surface area (1800 sq in/quart), does not shrink or clog, can be fit into any space desired, long maintenance intervals (6-24 months).
• Siporax® and other porous sinter glass media: very large surface, very expensive, crushes easily, i.e. short life when rinsed repeatedly, fine particles wear out powerhead rotors, sludge remains in the filter when lifted out for rinsing.
• Zeolith: absorbs ammonia, nitrates, and other potentially harmful waste products but may also absorb calcium and other nutrients, heavy, expensive, fine particles wear out powerhead rotors, inconsistent quality because it is a natural product.

The “perfect” filter

By considering all the information above, a filter and its material should have the following properties: large outer and inner surface, long maintenance intervals and lifespan, and the ability to provide good conditions for both nitrifiers and denitrifiers. Nature does that in soil layers of wetlands and in riparian zones (Nakasone et al 2003). Riparian (ripa means river bank in Latin) is the interface between land and a river or creek. Thus, by trying to mimic this habitat we can expect superior results to keep our aquarium water in good conditions.

Figure 1: HMF in theory (top) and practice (bottom). Click on the figure for a higher resolution picture

The Hamburg Mattenfilter (HMF) was invented several decades ago in Germany. The idea is to set up a vertical wall of Poret® filter foam on one end of the tank or across a corner with a little space behind (Figure 1). Jetlifter™ tubes or small power heads are used to pump water from the open space, over the foam, into the tank. The water then flows passively through the foam pores and gets filtered by the microorganisms that settle there. The filter foam becomes basically a vertical undergravel filter as long as it is large enough and left alone for many months. The latter is absolutely essential since any disturbance will affect the community of both nitrifiers and denitrifiers. Thus, the material needs to be strong enough to stand up for a long time and should not shrink or decay. With an HMF it is still possible to occasionally remove it for rinsing without completely taking apart the tank as with an undergravel filter, thus offering ease of maintenance. Moreover, it can be used in tanks that lack substrate, such as importer or breeding setups and growing tanks for fry, thereby essential taking over the function of the missing substrate.

While it is sometimes claimed that most of the filtration in an HMF only takes place in the first 1-2 cm (1 inch), the empirical evidence from filter sheets that are 10 cm (4 inches) or thicker shows that thicker sheets provide additional benefits since the accumulation of nitrate is slower. Presumably, this happens because the nitrification first reduces the oxygen levels, which allows more pockets of anaerobes to thrive and denitrification to take place much like in an undergravel filter or in the riparian zone. The volume of the water flowing through the space certainly affects how the community develops. However, it is possible that many anaerobes are facultative anaerobes and thus can adapt back and forth depending on conditions. As a rule of thumb, the higher the flow rate the thicker the foam should be.

In summary, freshwater biofiltration with its aspects of nitrification and denitrification is a largely unexplained area of research that we are only beginning to understand. Luckily, even without comprehending all the details, we can imitate nature by equipping the tanks with filtration methods that simulate microhabitats such as found in the riparian zones. The success of the simple HMF method in Europe over the past 25 years is now clearly supported by recent scientific understanding of what is going on in this bioreactor.

Disclosure

The author is the owner of Swisstropicals LLC, the exclusive importer and distributor for Poret® filter foam and B&H Jetlifters™ in North America. www.swisstropicals.com

References

• Nakasone, H, Kuroda, H, Kato, T, and Tabuchi, T (2003). Nitrogen removal from water containing high nitrate nitrogen in a paddy field (wetland). Water Science and Technology, vol. 48, no. 10, pp. 209–216.

• Sauder LA, Engel K, Stearns JC, Masella AP, Pawliszyn R, et al. (2011) Aquarium Nitrification Revisited: Thaumarchaeota Are the Dominant Ammonia Oxidizers in Freshwater Aquarium Biofilters. PLoS ONE 6(8): e23281.