To Skim or Not to Skim: That is the Question
 

by Scott Dyer and J. Charles Delbeek M.Sc.

Part 1: What is a Protein Skimmer?
If you were to ask a German aquarist whether he thinks protein
skimming is important he might look kind of puzzled and rhetorically
ask if you felt salt was important! A Dutch hobbyist might tell you
that it is a nice but unnecessary option, and a North American might
look dumbfounded and enquire as to what the heck you are talking
about!

Protein skimming may well be the most confused subject in the whole
hobby. From what I understand it was fairly popular in North America
ten to fifteen years ago, but for some reason this popularity steadily
declined. I have been an active hobbyists for about six years now and 
I when started keeping marines very few people (retailers or 
otherwise) could even tell me what protein skimming was supposed to
accomplish. 

Since the advent of the "Mini-Reef" and the resultant influx of
European information regarding the hobby, the interest in protein
skimming is once again on the upswing. In light of this I thought it
might be helpful if I shared what I have learned regarding this
subject over the past few years.

I remember the first time I saw a protein skimmer, it looked so hi-
tech and scientific that I wanted to buy it if for no other reason
than to impress visitors! In reality the functionality and use of a
protein skimmer is quite simple, even I had no trouble setting one up
for the first time. 

The skimmer consists of a contact tube, a collection dish, an airstone
and in some cases a water return tube. The water return tube is used
for reverse flow skimmers. Air is pumped through an airstone and rises
up the contact tube forming a foam. This is why it is sometimes
referred to as foam fractionation. This foam then rises through an
inverted funnel and is deposited in the collection cup. That's really
all there is to it.

Although a simple process mechanically, protein skimming gets a little
more complicated when we examine the chemical processes that make it
effective. Certain molecules in the aquarium are attracted to the
surface of air bubbles. These molecules are referred to as surface-
active molecules or surfactants. Copper, magnesium, manganese,
calcium, detritus and microalgae cells are all surfactant molecules. 

These organic molecules are polar at one end and non-polar at the
other end. The polar end is attracted to water molecules (hydrophilic)
and the non-polar end is repelled by water making it hydrophobic.

These surface active molecules are attracted to a surface where the
hydrophobic end of the molecule can be away from water and the
hydrophilic end can remain in contact with water. Air bubbles provide 
an excellent air-water interface for the surfactants to adhere. One 
side of the bubble is in contact with water for the hydrophilic end, 
while the inside of the bubble provides an environment (in this case 
air) for the hydrophobic end. This is why protein skimmers utilize an 
airstone or venturi to pump fine air bubbles up through a contact tube 
and into a collection cup.

There are two major factors that affect the efficiency of a protein
skimmer. They relate to the ability of air bubbles to attract and
discharge surfactants in the collection cup. They are 1) bubble size
and 2) contact time.

1) Bubble Size: In general it can be said that the smaller the
   bubbles the better. Smaller bubbles provide more surface area per
   cubic inch for the surfactants and rise through the water at a
   slower rate providing more contact time. The only time a problem
   could arise would be a situation where the bubbles were so small
   that they could not break the surface of the water and produce
   foam.

Spotte (1979) states that the more surfactants in the water, the
smaller the bubbles will be. This is caused by the effect
surfactants have on the surface tension of air bubbles. Under
normal circumstances the wooden airstones provided with most
skimmers are sufficient.

2) Contact Time: It is important that the air bubbles have as much
   contact time with the water in the column as possible. I have seen
   skimmers for large systems that were over six feet (2 m) tall! Of 
   course this is not practical for most home aquaria, but the contact 
   tube should be as large as possible none-the-less. Skimmers placed 
   in the aquarium or filter limit the contact tube to that depth. 
   This is why free standing skimmers used outside of the aquarium are
   preferable.

It is now time to have a look at the advantages of skimming over or
complimentary to other forms of filtration.

Before we do this we should first take a moment to examine just what
protein skimmers remove from solution. One of the best discussions I 
could find in English on this matter is from Peter Wilkens' book, The
Saltwater Aquarium For Tropical Marine Invertebrates. This book
contains one of the best available discussions on protein skimming
although it is slightly outdated (1973). Other good sources are Moe 
(1989) and Thiel (1988;1989). 

Protein molecules are not the only substances removed by skimming.
Tests have shown that larger particles such as green, red and brown
algae are also removed. Proteins are water-soluble molecules (ions)
and therefore react quite readily with other substances because of
their positive and negative charges (metal ions such as zinc and
copper are examples of such substances) resulting in large complexes. 

As most aquarists know, these substances are very harmful to algae
and invertebrates even at relatively low concentrations. There is,
however, one disadvantage to removing these ions. A good skimmer will
remove such minute amounts of these metals that not even traces will
remain in solution. It may be necessary therefore to replenish some of
the trace metals as time goes on. This should only be done after
testing the water with an accurate test kit to ensure that you don't
overdose the algae and animals. 

Other substances pulled out of solution with a protein skimmer include
phenols, phosphates and carbohydrates (Wilkens 1973). This brings us
to one of the most important advantages of using a protein skimmer:
large organic molecules are removed from solution before they are
broken down by mineralizing and nitrifying bacteria into ammonia. This 
reduces the load on your biological filter and raises the redox 
potential of the water. By the way, "redox potential" is simply the 
total potential of your water to either reduce or oxidize organic 
substances.

An important factor affecting the efficiency of your protein skimmer
is foam or "scum" formation as it is often called. Wilkens (1973) 
discusses two basic types of foam that develop in effective skimmers. 
The first is referred to as "standard" scum and the second as 
"protein" scum. It is the second layer, the protein scum that removes 
harmful organic substances from solution. If your skimmer only 
produces the standard scum it is not being used effectively.

Standard Scum: The standard scum is the layer of foam that develops as 
   a result of protein molecules uniting with fine air bubbles. Most      
   skimmers will easily extract this layer of foam. However, the      
   amount of harmful organic substances removed in this way is      
   minimal and may not curtail the development of "toxic tank      
   syndrome" etc. 

Protein Scum: Protein scum is formed by organic substances, metals, 
   etc. attaching themselves to the protein molecules in the      
   standard scum. Protein scum is therefore found as a second layer      
   beneath the standard scum in the column of the skimmer.

In light of this one can see that it is essential that your contact
tube be long enough to facilitate the development of both types of
scum. Wilkens (1973) states: "In a well adjusted skimmer, the protein 
scum will slowly rise in the column until it is collected in the foam-
cup". He also states: "One will get a proper adsorption and with that 
a concentration of the proteins on the edge between standard scum and
protein scum".

Another important factor in the development of both types of foam is
turbulence. If the air bubbles are swirling around in the contact
column and crashing up against the walls of the column, proper foam
formation will be greatly inhibited.


Part 2: Construction and Implementation of an External Protein Skimmer
[There is actually a better skimmer design elsewhere on this site]

The advantages of an external skimmer are many not the least of which 
is the removal of an unsightly piece of equipment from the aquarium. 
However, perhaps the most important advantage is that the size of the 
skimmer is no longer restricted by the size of the aquarium. This is a 
very important consideration because the larger the skimmer is, the 
greater the volume of water it can handle. In addition, the greater 
the length of the skimmer the greater the contact time between the 
water and the rising column of air bubbles, which is the limiting 
factor in the efficiency of any skimmer. The only real restriction on 
the length of the skimmer is the strength of your air pump since you 
have to pump the air down through that column of water.

Air Injection
Before I get into the design of the skimmer itself, I would like to 
discuss the method by which air is injected into it. In order for the 
skimmer to work effectively the air bubble should be less than 0.5 mm 
in diameter (Achterkamp 1986). The most common method to achieve this 
is to use a wooden air block. I have tried a number of different 
airstones over the years but for pure efficiency nothing can beat a 
wooden air block. True, they are expensive but they are the best! You 
can, however, make your own blocks. I've made a few out of 
white/silver birch; at $1.50 for a square foot you can get about 20 
blocks. You can experiment with other wood types if you like as long 
as they are porous enough (some people suggest balsa wood but it may 
contain too may pores to work effectively when wet). If you buy your 
blocks ready-made there is a way to recycle used blocks when their 
output drops. First you soak the blocks in a pure bleach solution such 
as Javex or Chlorox, for a couple of days. This performs two 
functions: it dissolves any calcium deposits and it swells the wood 
thereby reopening clogged pores. Next, soak the block in successive 
changes of freshwater to remove any hint of bleach then let the blocks 
dry for completely or dry them in a oven before reusing them. 
Following this method the block often works better and longer than 
before (Wayne Coll, personal communication).

There are, however, a number of problems with using air driven wooden 
blocks. First of all an air pump does not supply a constant air 
pressure over time, especially those used by the average hobbyist, so 
the rate of bubble formation varies considerably over time. Secondly, 
wooden air blocks are very susceptible to rotting which blocks many of 
the air pores. This results in a varying bubble size and rate of 
production, which ultimately results in a poorly functioning skimmer 
(Achterkamp 1986). An alternative method to the wooden air block-air 
pump combination is the use of an injector valve (Fig. 1). An injector 
valve consists of a small device which passes a thin stream of water 
through a very small opening. The water passes through this opening 
into a larger chamber which causes the water to expand rapidly; 
commonly known as the Venturi effect. There are small openings in this 
chamber leading to the outside. The rapid expansion of the water 
causes air to be sucked in through these openings, forming a very fine 
mixture of air and water, this mixture is then introduced into the 
bottom of the skimmer. The water used in the diffuser can be bled off 
from the main water supply to the skimmer (Fig. 1b). An injector valve 
rarely needs to be replaced and circumvents many of the problems 
inherent in wooden air blocks but they are not without their problems. 
To make the valve work properly, the water must enter under pressure, 
this requires that the water be pumped to the skimmer by a strong 
water pump. Some injector valves can be adjusted to give the optimum 
air bubble size and need to be regularly checked and adjusted during 
periodic cleaning of the skimmer. A skimmer which uses an injector is 
much more powerful and efficient than a normal skimmer and some argue 
that they can actually be hazardous to the health of your fish if run 
continuously, however, I have seen no documented proof of this. There 
are now a number of skimmers available on the market which utilize the 
Venturi principle and are commonly sold as Venturi skimmers. Most 
consist simply of a water inlet to which a T-fitting has been 
attached. A length of air line is attached to the fitting with an air 
valve located at the end. The air valve is used, in turn, to regulate 
the amount of air entering the skimmer and to adjust bubble size. 

Size of the Skimmer
The size of the skimmer you construct depends, of course, on the 
size of tank which you want filtered. I have seen commercial external 
skimmers ranging in height from 40 cm (16 in.) to 240 cm (8 ft.)! A 
skimmer with a diameter of 15 cm (6 in) and a height of 120 cm (4 ft) 
can handle aquariums up to 600 litres (160 gal) in capacity, using a 
flow rate of 1200 litres (315 gal) per hour (Achterkamp 1986). By 
using these values you can determine the dimensions the skimmer should 
be for your particular aquarium. I would think that a skimmer of 15 cm 
x 60 cm (2 ft) would be more than adequate for aquariums under 400 
litres (100 gals). Just remember, it is the height of the skimmer 
which is most important in determining the pressure required to run an 
airstone; the diameter is irrelevant (pressure = mass per unit area; P 
= m/a). You can put more than one wooden air diffuser in the skimmer 
provided you have enough room; those greater than 10 cm (4 in.) in 
diameter will work more efficiently with at least two blocks.

Construction
The external skimmer illustrated in figure 2a can be constructed from 
just about any non-toxic material. P.V.C. or acrylic tubing are 
probably 2 of the more common materials available. P.V.C. tubing is 
readily available in a variety of diameters and lengths, acrylic 
however, is more difficult to come by, most manufacturers require that 
you buy a minimum length and it is more expensive than P.V.C.. The one 
advantage of acrylic is that it is clear. This lets you easily check 
that the skimmer and air block are working properly and that the tube 
has not become fouled by slimy algae. You can purchase clear P.V.C. 
too but it is quite expensive. As mentioned previously the length and 
diameter of the tube is up to you, however, the tube should not have a 
diameter of less than 10 cm in order to install the wooden block and 
return tube properly. The cover is made from two circular sections of 
acrylic, one with a diameter greater than the main tube of the skimmer 
and one with a slightly smaller diameter than the INSIDE of the main 
tube, and a rubber O-ring (Fig. 2b). In the middle of each plate a 
circular hole is drilled to accommodate the foam collection tube. A 
smaller hole is drilled near the edge of each plate to accommodate the 
rigid airline tubing for the airstone. For the foam collection tube 
you could use some tubing from an old Sander internal skimmer which 
had an outside diameter of 5 cm. If the tube fits snugly there is no 
real need to glue it in, this makes it easier to clean later. You can 
use also the top half of the old Sander skimmer which contains the 
foam stabilizing grid and holds the collection cup. A foam 
stabilization grid could be added by gluing two sections of egg 
crating into the foam collection tube, one on top of the other but 
this is an option. Next, 3 acrylic buttresses should be glued around 
the outer circumference of the top of the main tube in a triad 
pattern. These buttresses can have any shape you like but should be 
predrilled and threaded to accept the nylon locking screws. Once the 
buttresses are in place, put the larger of the two covers on top of 
the main tube and mark the position of the buttresses on this plate 
then drill holes in the plate to accept the screws. Once this is done 
the two circular plates are glued together with any type of acrylic 
glue. I would recommend that the inner cover be at least 1/4 inch 
thick, 1/2 inch would be even better. When you buy the O-ring check 
the thickness, this is important when calculating the diameter of the 
inner circular plate. What you want to end up with is a plate which 
fits snugly into the main tube of the skimmer such that the O-ring is 
somewhat flattened against the tube making a good seal. If you make 
the inner cover too small the O-ring will not press against the side 
of the skimmer firmly enough and it will leak. This method of sealing 
the skimmer is the same as used in Eheim and Fluval power filters. You 
can, of course, use any method you like to make the top and fasten it. 
What I have shown is only one way of doing it, I'm sure some of you 
could think of better ways, please feel free to experiment, that's 
part of the fun of being in the hobby!

Once you have the top of the skimmer finished insert the cover and 
mark the lowest point that it reaches into the tube, on the outside of 
the main tube. This line marks the upper limit of where we will be 
drilling the next set of holes; the inlet and outlet holes. The upper 
edges of these holes should be at least 1.0 cm (1/2 in) below the line you 
drew. It doesn't matter where you put the holes in relation to each 
other, as long as they are on the same level (Fig. 2a). They can be 
side-by-side, 90 degrees or 180 degrees from each other as long as 
they are at the same height, that's all that matters. The inlet hole 
should be smaller than the outlet hole (e.g. inlet dia.= 2.0 cm (3/4 
in), outlet dia.= 3.0 cm (1 in), this ensures that the water will flow 
out of the skimmer more easily and not out the top. Once you have 
drilled your holes, you can glue in the fittings. These can be easily 
purchased from any hardware store or from mail order companies 
advertising in aquarium magazines. The inlet fitting can be a straight 
through fitting or an elbow while the outlet fitting should be an 
elbow (Fig. 2a). Once the fittings have been glued in and the glue 
hardened, you can add a piece of rigid tubing to the outlet tube so 
that it reaches to a point below where you expect the airstone to be. 
You are now ready to add the bottom. This can simply be a piece of 
acrylic of any size as long as it covers the bottom and extends out 
from the sides of the main tube to provide support for the skimmer so 
that it doesn't topple over. Once everything has set and dried, and 
you have tested the system for leaks, the skimmer will be ready for 
installation. 

Installing the Skimmer
There are two basic methods for putting your skimmer in-line with your 
filtration system be it a dry-wet filter or a conventional system. The 
first method is to divert some of the water traveling from the 
aquarium to your filter, to the skimmer and then back into the filter 
(you can also add it to the sprinkler tube of a dry filter so that no 
water by-passes the dry filter) (Fig. 3a). A control valve is placed 
between the water supply and the skimmer to control the flow of water 
into the skimmer (Fig. 2a and 3). An additional valve should be placed 
between the skimmer return and the filter. This allows for a fine 
adjustment of the flow rate through the skimmer. The second method is 
to place the skimmer AFTER the filter (Fig. 3b). In configuration B 
the return water can be passed back to the filter or it can be added 
directly to the aquarium, however, in the latter case, the skimmer 
outlet tube must be slightly higher than the water level in the 
aquarium to ensure that the water will flow to the tank and not out 
the top of the skimmer! Both configurations A and B will work fine, 
however, I feel that configuration A is the better of the two in terms 
of performance. By taking "raw" water from the aquarium and skimming 
it first, you will remove a lot of the organic molecules and colloids 
before they reach your filter. If you are using Chemipure or activated 
carbon, the use of a skimmer located before them will greatly prolong 
their operating life. 

The Skimmer in Operation
Once you have the skimmer connected in-line, connect the air supply to 
your air blocks and turn on the pump THEN open the control valve 
slightly to let the water into the skimmer. This prevents the wooden 
air block from getting wet; its much harder to start a wet wooden 
block under 60cm (2 ft) of water than a dry block! You will have to 
watch your skimmer closely at first to make sure that the flow is just 
right, too great a flow will result in a full skimmer, foam tube and 
collection cup, a wet floor and an empty aquarium! To guard against 
such a calamity, you could drill a hole near the top of the collection 
cup and install a drainage tube which would lead back to the filter. 
Don't make the drainage hole too low in the cup or else the stuff that 
we have worked so hard to remove will just end up back in your 
aquarium! In the case of a blocked return tube or a 2 year old playing 
with your control valve, the excess water will just flow back into the 
filter/aquarium. One last word of warning: Never allow the skimmed off 
material to flow back into the aquarium or you could be left with a 
deadly toxic situation.

You should notice a dark brown sludge accumulating in the collection 
cup after only a few hours, longer depending on the state of your 
water. During this period you might have to empty the collection cup 
2-3 times a day, but the output will soon slow down appreciably until 
it stops all together. The skimmer is still working, there just isn't 
much left to remove that's all. You'll notice that the skimmer will 
begin to work furiously after feedings, water changes, addition of 
trace elements or the addition of fresh live rock (boy will it ever 
foam after you put in live rock!).

Now that you've set-up your skimmer you're probably thinking "At last 
I'm finished", well not quite, you see in order for your skimmer to 
keep working at peak efficiency it must be kept clean. A monthly 
cleaning is all that is usually required. This involves cleaning out 
the foam collection tube, the foam stabilizing grid (if you have one), 
scrubbing the inside of the main tube and checking the state of your 
wooden air blocks. The inside of the skimmer must be kept clear of 
algae and any build up of organic material on the sides. A bottle 
brush works fine for the main tube and foam collection tube while a 
coffee percolator brush works great on the stabilization grid.

Using Ozone
The use of ozone in reef tanks, as a method of raising redox 
potential, has been gaining support in the last few years. If you plan 
on using ozone make sure that all the materials you used in building 
your skimmer are ozone resistant. This applies especially to the O-
ring, it must be made of silicon. If you plan on using ozone in your 
skimmer you must also make allowances for residual ozone returning to 
the tank and atmosphere. Your collection cup should be sealed at the 
top with a small opening to let air/ozone escape. Over this opening 
you can place a small container of activated carbon to absorb residual 
ozone. In your water return from the skimmer, you must place also a 
container of activated carbon, such that all of the water has to pass 
through the carbon before it returns to the aquarium. This will remove 
any excess ozone and other by-products of ozonation which may have 
been produced.

Finally, the last question one usually asks is "is a skimmer really a 
necessary piece of equipment?". In our opinion, skimmers are very 
useful pieces of equipment for ANY marine tank, not just reef tanks, 
and should be used whenever possible.

Trouble Shooting

Q. Why won't my skimmer foam!?

A. There are a number of possibilities to explain why a skimmer won't
   foam and many of them are related to the theory of how a skimmer
   works and what affects its efficiency:

 1. If your tank has just been set-up with new water there is a
    strong possibility that there is not yet enough organic matter
    present to cause the skimmer to foam.

 2. The air bubbles are too big. The air bubbles must be < 0.5 mm in
    diameter.

 3. Your skimmer is too short and/or the contact time of the water
    with the air bubbles is too short. If the contact time is too
    short there will not be enough time for the air bubbles to
    collect enough organic molecules to cause foaming. The cause of
    a short contact time can be either; 1) a skimmer which is too
    short; 2) a flow rate which is too great, therefore your skimmer 
    should be wider or your flow rate should be slowed and /or; 3) a 
    short skimmer and a high flow rate.

 4. Your skimmer is located after another mode of chemical
    filtration (i.e. activated carbon, Chemipure, Polyfilter, etc.) 
    and this medium is removing the majority of the organic wastes 
    before the skimmer can receive them.

 5. If you are using ozone you are either adding too much or not
    enough. You'll probably have to fiddle with the ozone output to
    get the skimmer to foam.

 6. Fats or oils have been introduced either from foods or your skin.
    Everytime you put your hands in the tank you are introducing oils 
    into the water. These oils lower the surface tension of the water 
    which inhibits foaming.

Q. My marine plants and invertebrates seem to be on the decline, what
   should I do?

A. An actively functioning skimmer is a very efficient beast and may
   remove trace elements from your aquarium to the point where the
   aquarium inhabitants, especially the plants, will begin to suffer.
   There are two courses of action you can take and probably a
   combination is best. First you can limit the number of hours you
   run the skimmer. If your skimmer only foams after feedings then
   this should be when you run it. During other periods you can either
   shut off the flow valve or turn off the air pump. If you spent
   hours adjusting the flow you may not want to shut off your valve
   but if you shut off the air for long periods of time water will
   penetrate the wood, speeding-up the rotting process. A completely
   soaked wooden block is tough to restart with the majority of
   commercial air pumps available. The other solution is to regularly
   add trace elements and keep an eye on their levels by using a test
   kit such as the one for iron sold by Dupla or LaMotte. Combisan 
   sells a trace element solution, developed by Peter Wilkens, which 
   comes complete with directions on the amount to be added depending 
   on the size of your skimmer and how often you run it. Personally, I 
   feel that if the skimmer is not foaming then nothing is being 
   removed, so why shut it off? There is still a great deal of debate 
   as to whether a skimmer removes a significant amount of trace 
   elements and whether or not they are replaced naturally by the 
   metabolic actions of the animals and plants as well as the bacteria 
   found in live rock (Achterkamp 1986; Keith 1980).

References
Achterkamp, A. 1986. De eiwitafschuimer ... ja-ne? Het Zee-aquarium
                     36 (3): 55-60.

Keith, R.E. 1980. Protein skimmers in the marine aquarium. F.A.M.A.
                     vol. 3, no. 9: 20-21.

Wilkens, P. 1973. The Saltwater Aquarium for Tropical Marine
                  Invertebrates. Uppertal Elberfeld.

Recommended Readings
Moe, M.A. 1989. The Marine Aquarium Reference: Systems and 
                Invertebrates. Green Turtle Publ., Plantation, FL.

Spotte, S. 1979. Seawater Aquariums: The Captive Environment. Wiley-
                 Interscience.

Thiel, A. 1988. The Marine Fish and Invert Reef Aquarium. Aardvark 
                 Press, Bridgeport, CT, USA.

--------- 1989. Advanced Reef Keeping I. Aardvark Press, Conn.