Of the 3 types of filtration, i.e., mechanical(physical), biological and chemical, biological is by far the most important filter to have in the tank.


Mechanical Filtration - use of media like sponges, to filter off solid particles like solid waste, uneaten food, detritus.

Biological Filtration - use of beneficial bacteria to break down products in the nitrogen cycle like NH3, NO2-, NO3-.

Chemical Filtration
- use of chemical resins and carbon to absorb specific ions like NO3-, PO4-, colours, smells, etc.
A protein skimmer usually falls into this category even though it's using air as a basis of enabling organic particles to stick to the foam.

Over the past decades, many filters have been designed for FW and SW tanks, using a wide variety of media.
There are the undergravel filters (UGF), box or corner filters, hang-on-back (HOB) filters, canister filters like the Eheim, internal filters attached to the bottom of a powerhead, internal filters that are complete units with mech., bio. & chem. sections, overhead filters, fluidised bed filters, UV-filters, and numerous others.

There is more than one way to skin a cat, which implies that no one filter is perfect for everything.
However, certain filters like the UGF are generally not acceptable in marine tanks because of
detritus build-up in the gravel/crushed coral bed and clogging of the filter plates.

You'll just have to keep telling yourself time and time again : "Think Biological, BUT think carefully".

Bio-filtration relies on the use of various types of beneficial bacteria in order to
process/breakdown/assimilate/mineralise organic wastes to
ammonia NH3 (toxic),
to nitrites NO2- (toxic),
to nitrates NO3- (non-toxic in low concentrations, very problematic in high conc.),
and finally to nitrogen gas, N2 (totally inert and harmless).

When you reach N2 in gaseous form that gets released into the water, your nitrogen cycle is complete.
The beneficial bacteria is called Nitrobacter and nitrosomonas, but the names are not too important.

Just remember that there are 2 main phases in the nitrogen cycle :

1) nitrification, handled by nitrifying bacteria in an aerobic (with oxygen) environment.
2) DEnitrification, handled by denitrifying bacteria in an anaerobic (no oxygen) environment.

The nitrification phase handles the "conversion" of NH3 to NO2- and NO2- to NO3-.

The 1st bacteria that does NH3 to NO2- multiply pretty quickly, usually within several hours,
but may not completely convert the NH3 quick enough, depending on how high the level is.
This of course depends on how soon you put in the fish after you set the tank up,
and also how much you've been TRYING to feed the fish.

The 2nd bacteria that does NO2- to NO3- multiplies much slower than that,
possibly taking as long as 2 weeks to have enough to handle the NO2- levels produced by bacteria 1.
All filters with biological media will handle this part of the nitrogen cycle pretty well, with some being faster than others.
The highest speed ones seem to be the trickle filters and the fluidised bed filters.
However, the faster NO2- is broken down, the faster NO3- is produced!!
So sometimes this may not be too desirable.

This method is only good if you need to set up a tank in a hurry, say for an exhibition,
and you don't mind changing the water after the show. Or relocating the fish to a safer and stabler environment after that.
It is in cases like this that aquarists will jump-start the nitrogen cycle by using things like BioZyme, Cycle, etc.

The denitrification phase is the slowest and usually the phase that cannot be achieved by most of the filters mentioned above.
The reason for this is that in traditional filters, there's lots of water flowing thru' the filter media,
and this water carries lots of oxygen ... if it didn't, your fishes and corals would probably already be dead.
Unfortunately, in the presence of oxygen, the denitrif. bacteria cannot survive,
and thus the NO3- that are produced by the regular filter media cannot be converted to N2 gas,
and this NO3- level will climb higher and higher until it's well beyond control.

Problems with Nitrates : Fish can do well at levels of up to about 40ppm NO3- easily,
and I've seen some tanks with NO3- levels at 400 ppm!!!
At these elevated levels, fish growth will have stopped, they have internal problems,
and your tank will most likely be totally covered with a disgusting layer of red slime algae,
commonly known as cyanobacteria.

No fish eats this type of algae. Period! It's quite easy to siphon out the algae, since it's very light and doesn't stick hard to the surfaces.
However, even if you siphon out everything today, in 2 - 3 days, it will be back even faster than before.
There is no other way to go about solving this problem other than tearing down the tank and resetting it up then.
Not gonna be fun

This is the reason why people like Schott invented Siporax, a filter media using something called sintered glass (haven't figured out what it is yet!!),
which is structurally diff. from crushed corals, bio-rings, etc. The inner portion of sintered glass has a lot of very fine micro-pores,
which do not let much, if any water pass thru'. As a result, the denitrifying bacteria can live in here.
Unfortunately, Siporax is VERY expensive, and still not as efficient as a deep sand bed, as will be detailed further down below.

The Berlin system and DSBs : Sometime in the early to mid- '90s, some people in the US realized that by using LR (live rock),
a protein skimmer and nothing else, their NO3- levels could be kept very low, much lower than what was previously possible with any other filter,
EVEN with Siporax, provided they used enough of it.

They discovered that deep within the inner portion of a piece of LR lived lots of denitrif. bacteria since water doesn't really flow thru' it much.
The protein skimmer exported most of the organic wastes BEFORE they became NH3,
so it gave the bacteria colonies a headstart.
For the amount of LR to use, normally we put in about 0.6 - 0.7kg per US gallon of water.
In our case study tank of about 15 USG, I'd put between 9-11kg of LR,
right on top of the sand bed that was made in Marine Tank Initial Setup.

Towards the later part of the '90s, deep sand beds (DSB) started getting being used in reef tanks.
And people then realized that 0 ppm NO3- was actually possible.
This was because when you're using oolitic aragonite, or #0 crushed corals, or even silica sand,
there were so many sand particles crammed next to each other that the surface area for bacterial colonisation was fantastic.
Added to that was that the deeper you went in the sand bed, the less oxygen there was, so beyond the first 1.5" or so of the DSB,
all the below portions were mostly anaerobic!!!

Basically, a DSB is just that. Nothing more than lots of sand at the bottom of the tank, with nothing underneath it.
Most DSB depths range from 2" to 6", depending on how tall your tank is, how much swim room you need for your fish,
and how heavy a bio-load you anticipate to have in future.

For our case study tank, we can afford to have 3" of DSB, and still have about 11" of vertical swimming space,
and 1" of air space left behind. Most fine sand or aragonite weigh in at between 32kg - 36kg per cu.ft.,
depending on the pore size. In order to calculate the weight of sand needed,
you can do L(feet) X W(feet) X H(inches/12) X 35kg to get 17.5kg of #0 sand.

Tank cycling : For new setups in order to avoid sudden and unexplained deaths
caused by New Tank Syndrome (NTS), the tank has to cycle.
What this means is nothing more than giving it enough time for all 3 types of bacteria to multiply and colonise the sand bed,
assuming that your LR is already colonised.

Generally, this takes anything up to a month, and its success can be observed easily
by examining the DSB after the 1st week or so.

As time goes on, usually after the 1st 2 weeks, you'll notice more and more little "air" pockets forming in the DSB next to the glass.

This is not air, but nitrogen gas which indicates that the denitrifying bacteria have already started multiplying.
Remember to test your water for NH3, NO2- and NO3- on a weekly basis,
which will let you know the status of your tank's stability - you CAN'T guess.

After a month, NH3 and NO2- should both be zero.
NO3- may or may not be - usually it'll be low, but not quite there yet.
If it's less than say, 10ppm, you can go ahead and put in some easy corals,
but don't put any fish in yet until the NO3- have actually reached zero.
This will give the colonies a real headstart against what fish you add later.
Better still to wait for it to have reached zero first.

Reason why you can't have as many fish in a marine tank as you can in a FW tank :
Even when your tank has cycled, you should never overcrowd your tank for 2 reasons :

1) Overcrowding causes stress, which can then cause infections to start in your fish.

2) Higher pH levels than in FW means that most NH4+ (ammonium ions) are converted to NH3 gas
which is many times more toxic than the ion found in FW tanks (unless you're talking about Malawi or Tang. tanks).

Other regular water tests to do : KH
- helps to buffer pH (much more than crushed corals) and prevent it from swinging,
as long as it's kept at 8 degrees or higher.
- Also contributes largely to formation of calcium carbonate in coralline algae and coral skeletons.

Calcium - especially necessary if you plan to keep hard corals,
but also good to know, in order for your coralline algae to grow quickly.

Final note : You can still use sponges, but not on a regular basis because
they will filter off the plankton in your tank produced from the shrimps, crabs, worms, snails, etc., in the LR and DSB,
which will deprive your corals of their natural food source. Carbon used long term will deplete your trace elements.

Once in a while however, you may come across a situation where Berlin plus DSB may just not be good enough for the application.
A typical case would be that of a predator tank.
All meat eaters generate lots of waste, and NH3, NO2- and NO3-. It may be and usually is too much for these filters to handle.

What we normally do in this case is to use trickle filters coupled with denitrifiers,
or skip the denitrifier and do more regular water changes of larger amounts.

You can still have the LR and protein skimmer and DSB, and this will help to a certain extent, but not on their own.


Setting up marine aquarium main

Back to aquarium information main


All Contents Copyright 2000-2003 All Rights Reserved.