Toxicity Of Saltwater Aquarium Trace Elements

by | May 15, 2003 | 0 comments

In a study conducted by Atkinson and Bingman [1] it was shown that commercial salt mixes contain elevated concentrations of heavy metals compared to natural seawater. Studies of aquarium water by Salifert [2] and Shimek [3] have shown a similar trend. It has been stated by Shimek that the concentration of some of these elements as present in aquariums is toxic to many invertebrates including corals [4].

In this article I will explain that it is not always necessary that a heavy metal is toxic since this depends strongly on the form (=speciation) in which it is present. That is it is not always bio-available [5-14].

To demonstrate the limited bio-availability of at least some heavy metals in closed systems such as aquaria, analyses of coral fragments grown in natural seawater and in closed systems with elevated heavy metal concentrations are shown.

Various studies have shown that normally the skeleton of corals can be used as a proxy for heavy metal pollution [15-25]

We can namely expect a higher concentration of a metal being deposited in the corals’ skeleton with increasing concentration and bio-availability of the same metal in the water column. But also the opposite is true.

Speciation and Toxicity

First I would like to start with a non-aquarium example to explain what speciation means and how it can impact toxicity. Cyanide is considered to be highly toxic to humans but also to many other organisms. It is toxic because cyanide binds to the iron contained in the red blood cell’s haemoglobin. This iron in haemoglobin allows transport of oxygen by binding oxygen to the iron ion. Since cyanide also binds to the haemoglobin’s iron but with that difference that it is bound much stronger to the iron compared to oxygen, it inhibits the transport of oxygen. If there has been enough cyanide to block enough haemoglobin death will occur.

A very good antidote for cyanide poisoning is the administration of an iron salt. Cyanide will then react with this iron and not with the haemoglobin’s iron thus reducing the toxicity of the cyanide.

The free ionic cyanide is thus toxic, but the iron cyanides in general are not. There are many forms of iron cyanides and hexacyanoferrate is an example for this.

We also say that cyanide can have different forms or speciation and the term speciation will be used almost exclusively in this article from now on. Later on some more examples of speciation and toxicity will be given.

Something very important to know is that if a substance is analysed for cyanide we can not always conclude anything about it’s toxicity from such data because first we also need to know how much of each cyanide speciations are present. After all it could have been present as the toxic ionic cyanide, as the non toxic cyanoferrate, a combination of both or even a totally different speciation. So if a method of analyses is used and no distinction is made regarding speciation or is not determined by other means then the results are very likely of not much value if the intention is to conclude anything about toxicity. Saying that it is toxic would just be a mere speculation.

To continue with cyanide: did you know that table salt can contain significant amounts of cyanide? In many parts of the world hexacyanoferrates [26, 27] are added to table salt as an anti-caking agent. But since the cyanide is already very strongly bound to the iron (ferrate stands for the iron) it is no longer toxic.

So analyses for cyanide might show table salt to be harmful while in fact it is not. This again emphasises how important it is to know the speciation. So, if someone would say that table-salt is toxic or harmful because of cyanide then you are already warned. It is then just a speculation by someone who does not now that speciation can affect bio-availability and toxicity.

Now I would like to give an aquarium related example. This one does not deal with heavy metals but puts the terms toxicity and speciation even better in the picture because most aquarists already know that toxicity depends on speciation but many probably don’t know that they already knew it.

Ammonia, Nitrite, Nitrate and Nitrogen

What have ammonia, nitrite, nitrate and nitrogen in common? Well several things.

  • All are different speciations (forms) of nitrogen.
  • All can occur in aquaria
  • All can be transformed by bacteria.

We all know, speaking in aquarium hobby terms, that ammonia is about as or far more toxic than nitrite and that nitrate is far less toxic than ammonia or nitrite. Nitrogen gas is not toxic and air contains approx. 80% nitrogen gas.

Again all are different speciations of nitrogen, but their (toxicological) properties are very different. Some other speciations of nitrogen are amino acids, proteins and various amines.

It is possible to determine the total nitrogen content of a sample. In such cases the nitrogen gas is not determined. Such an analyses then gives the total nitrogen concentration as present in ammonia, nitrite, nitrate, amines, amino acids, and proteins.

Does an analysis for total nitrogen say anything about toxicity? No, it does not!

For this same reason we determine ammonia, nitrite and nitrate as different parameters — different speciations. Only then we might determine if the water is toxic or not.

With this example it has been shown that aquarists have already known for a very long time that toxicity depends on speciation and is nothing new to them, except for the term speciation.

This is known by aquarists because it is a subject which can be found in most if not all books dealing with aquarium keeping.

However, speciation of heavy metals and toxicity has not been dealt with often in aquarium books with copper the exception. Copper is a heavy metal and heavy metals are what we will be talking about.

Treating Parasites With Copper And Its Speciation

It is a common practice to treat fish suffering from some external parasite with copper. There are various ways (speciation) to do so. First is the addition of ionic copper as copper sulfate. It is known that this does not work quite as well because the required copper is readily adsorbed on rock and sand surfaces, is transformed to insoluble carbonates or gets bound to some organic compounds making it non bio-available and thus reducing it’s toxicity. After all, toxicity is required to kill the parasites.

Adding some citrate to the copper allows the copper to remain longer in the water column as the citrate and since the copper citrate is directly or indirectly bio-available it retains it’s toxicity for a prolonged period. Also instead of citrate some amines are often used and function in the same way as the citrate does.

With the above example it is shown that the speciation of the heavy metal copper, added as an ion, changes and becomes far less toxic or perhaps even totally non-toxic in just a day or so.

The term bio-available has been used a few times – it actually means that it can be absorbed and can be utilized by the organism. To explain this a bit more the cyanide example can be used for this purpose. The non-toxic cyanoferrate will be absorbed, but neither the iron nor the cyanide can be utilized because they are not split in our bodies and is excreted as the cyanoferrate. So the cyanoferrate is not bio-available. Otherwise it still would be toxic. So, absorption still does not say anything about the degree of utility — whether it’s beneficial or toxic.

Before continuing with this article a few things will be summarized.

Elements or substances can occur in different forms which is called the speciation. An analysis of metals giving the total concentration doesn’t yield information about its speciation. The toxic effects, if any, depend on bio- availability, which in turn depends on speciation.

Many organisms can change the speciation of metals thus changing its bio- availability. This is explained in the next section.

Speciation changing by organisms

Organisms including bacteria and algae are capable of changing the speciation of many substances. They can do this in their cells or even outside their cells; that is, in the water column. So they can by changing the speciation make some substances bio-available when they were not. But they can also reduce the toxicity of some substances by changing its speciation in a way which results in a highly reduced bio-availability.

Iron is an example of an element which is rather scarce in the water column because iron is readily transformed to the non bio-available iron oxide (rust) which has an extremely low solubility. Algae and bacteria require a relatively large amount of iron and to obtain this they excrete organic substances in the water which dissolve the iron oxide and bind the iron strongly to it. This iron is then bio-available. It is absorbed and enzymes break the iron (organic compound bond) and the iron is utilized. That is at least one of the pathways.

These ligands which they secrete are called siderophores and phytochelatins when excreted by bacteria and algae respectively. These are actually classes of substances and many or perhaps most of them have not been thoroughly understood yet.

When certain organisms are exposed to a concentration far beyond being beneficial and that substance is bio available and toxic at such elevated concentration then these organisms will produce some proteins called metallothioneins. These bind the metal in question making it no longer bio- available. Such substances are also excreted in the water column by at least some organisms such as algae and bacteria. If that happens then the substance will still be in the water column, but will no longer be bio-available and has then lost its direct toxicity.

This happens e.g., when too much ionic copper is added or was already present in a system which also contains bacteria or algae. Our tanks are a perfect example for such a system. So what we see is a dynamic response to variations in water chemistry, especially with respect to heavy metal concentration and its speciation in the water column.

Another interesting example to demonstrate how well organisms can respond and handle various water conditions is the excretion of an enzyme (alkaline phosphatase) by algae and bacteria to transform organic phosphate esters into inorganic phosphate when the concentration of the latter gets too low in the water column. These organisms do this because they do need inorganic phosphate and the organic phosphate esters are not bio-available.

Typical Speciations Of Heavy Metals In Seawater

Below are some classes of speciations. Within a class a heavy metal can either be bio-available or not, but that depends on many other factors which are far beyond the scope of this article.


Examples of ionic speciation are free and completely dissolved copper, zinc, iron, etc. Ionic forms are generally bio-available and can be toxic when the concentration is high enough. At elevated concentration compared to natural sea water, but below the level at which they become toxic, they can cause severe stress. Still, many (heavy) metals are essential for all organisms including corals, bacteria and algae. They are often the active site of enzymes or have another function such as the iron in haemoglobin. Also, this is perhaps the reason why organisms can make the same metal either bio- available or do just the opposite depending on needs and concentration.

Organically Bound

When the ionic forms are bound to organic molecules and that bond is not of a chemical nature then they are very likely chelated. The organic molecule to which they are bound, or perhaps it is better to say very strongly attracted, are called ligands. Examples of ligands which can occur in organisms in aquarium water and in natural seawater are humic acids, alginic acid, various other carbohydrates, organic acids such as citric acid, amino acids, and proteins.


The heavy metals can also be present as precipitated particles. These particles can be small enough to remain undetected by even the most powerful optical microscopes. These particles can be for example ferric oxide (rust), copper carbonate, copper hydroxide, zinc carbonate and copper sulphide.

It is also possible that ionic heavy metals are adsorbed on surfaces of calcium carbonate particles or on surfaces of detritus, algae, bacteria, and organic colloids.


Micro-organisms such as algae are well known to concentrate (bioaccumulation) trace elements such as zinc. So algae, but also bacteria and zoo plankton can contain relatively high concentrations of trace elements [28-29]. A bioaccumulation factor of 1000 or 10,000 times is nothing special. Organisms feeding on such micro-organisms can consequently also contain elevated concentration of such metals (biomagnification). More general information about bioaccumulation and biomagnification can be found at this site.

This has nothing to do with pollution, but is something which happens every day around coral reefs and in many other parts of the oceans. In analyses of various foods (done by Shimek [30]) often referred to by aquarists you will note the high copper content and even higher zinc content. The speciation of these metals in natural foods is, a priori, very likely to be such that they are, at least initially, virtually not bio-available.

Analyses Of Heavy Metals

Analyses of metals in seawater are often done with techniques such as ICP, ICP-MS and AAS. These techniques determine the total concentration of a given element. So they don’t give any information about speciation and to obtain information regarding speciation totally different methods are used.

If the sample is not filtered then heavy metals present as particles or adsorbed on it will be counted as well — also counted are the metals present in the floating algae and bacteria.

Note, the analyses mentioned in the beginning of this article and presented in studies by Atkinson and Bingman and Shimek used techniques which determined the total concentration of various metals. The samples were not pre-filtered.

Also the data which will be presented in this article were determined by a technique measuring the total concentration and does not give any information about the speciation. The technique used for the data presented here was ICP- MS.

All reagents used to dissolve the coral fragments and acidify water samples and water used for dilution were beforehand checked for contaminants, and was also the exact batch as used by that laboratory to conduct analyses on sub-ppb scale.


Coral fragments from corals growing in the wild were taken and had grown further in the open aquarium system of the Waikiki aquarium which has a continuous inflow of natural seawater. Fragments from these colonies were grown further in normal closed system aquariums.

Fragments of the same specie of coral are genetically identical. The only exception is the C. furcata of which no fragment of the mother colony was available, and instead of this a non-genetically identical fragment grown in the wild was used. However, the fragments of this coral grown in the Waikiki aquarium and in aquarium AA are genetically identical.

A. latistella was probably the first SPS grown in an aquarium and was found on live rock by Stuber in Berlin, Germany. The fragments of this coral are from his aquarium and from another aquarium in which a genetically identical fragment was grown.

All corals have been growing for at least 4 years in the various aquaria at the moment of sampling.

The fragments have been sun-dried. Unfortunately, some tissue was still present. This tissue can skew the results of especially barium (not shown here) and some heavy metals in such a way that an elevated concentration is found and does not reflect the true skeletal composition. Use of chemical methods to remove the small amounts of tissue was avoided since that could have caused a lowering of the concentrations in the skeletons. I would like to emphasise that any tissue if present can not significantly lower the actual concentration.


The results for copper and zinc are given in ppm for the corals and in ppb ppm) for the water.

This has been done intentionally. If the ratio of a heavy metal to calcium in the water is incorporated exactly as such then 1 ppb of the heavy metal in the water will show up (by approximation) as 1 ppm in the coral skeleton if the water’s calcium concentration was 400 ppm.

Table 1. Copper in water and coral fragments. Values for fragments are in ppm and for the water are in ppb.
CoralEnvironmentCopper in CoralCopper in Water
Caulastraea furcataWild0.8< 1
IbidWaikiki1.5< 1
IbidAquar. AA< 0.0115
Acropora microphtalmaWild1.1< 1
IbidWaikiki< 0.01< 1
IbidAquar. AA< 0.0115
Montipora digitataWild2.0< 1
IbidWaikiki< 0.01< 1
IbidAquar. BB2.3106
Acropora latistellaAquar. AA< 0.0115
IbidAquar. CC< 0.0118

The above results in general show that the coral fragments grown in closed system aquariums have a far lower copper concentration compared to their mother colonies grown in the wild, and this is in contrast to the typically 30 times higher copper concentration in the closed systems compared to natural seawater. Typically, natural seawater contains not more than 0.5 ppb copper.

The closed system Aquarium BB has a very high copper concentration and is approx. 200 times that found in natural seawater. Despite this the coral fragment of M. digitata has about the same copper content as its mother colony grown in the wild. Also, if all that copper would have been bio- available all the corals and many other invertebrates might have been at least severely stressed, especially since that situation lasted for at least one year.This ionic copper concentration (100 ppb = 0.1 ppm) is by many considered as lethal to inverts including corals and is about the same concentration used to treat fish diseases. A much lower ionic copper concentration has also been considered to have significant negative effects on zooxanthellae [31].

The above data suggest that almost all of the copper in the above closed system has not been bio-available for at least the examined species.

Table 2. Zinc in water and coral fragments. Values for fragments are in ppm and for the water are in ppb.
CoralEnvironmentZinc in CoralZinc in Water
Caulastraea furcataWild2.2< 1
IbidWaikiki1.5< 1
IbidAquar. AA1931
Acropora microphtalmaWild0.5< 1
IbidWaikiki2.1< 1
IbidAquar. AA< 0.0131
Montipora digitataWild8.9< 1
IbidWaikiki243< 1
IbidAquar. BB25432
Acropora latistellaAquar AA2.931
IbidAquar CC26< 1

It is striking to see the extremely high zinc content of the M. Digitata grown in the Waikiki aquarium. It is about 25 times higher than its mother colony grown in the wild.

It is quite well possible that the elevated zinc concentration is caused by ingestion of phytoplankton or phytoplankton feeders. Also, the presence of some tissue might have caused this since zinc is an essential part of Carbonic Anhydrase, which is needed for calcification and might be present in a high concentration in the coral tissue. Another possibility is the entrapment of detrital material. This coral sample shows that abnormally high values for at least zinc can be found even if the water is very low in zinc (natural seawater).

Given the fact that feeding but also contamination by some coral tissue or entrapped detritus or bacteria in the skeleton can result in an apparent higher zinc concentration in the skeleton analyses, the high values found for some skeletons should be viewed with some caution.

Nevertheless we still can say some things about bio-availability of zinc in some of the systems.

The fragments of A. Microphtalma and A. Latistella and both grown in Aquarium AA have a zinc concentration of < 0.01 and 6.4 ppm respectively. The zinc concentration in the water column is 31 ppm and approx. 50 times higher than natural seawater (which contains usually not more than 0.6 ppb). If this zinc would have been bio-available then we would have expected a far higher zinc concentration in these fragments. Since this is not the case it seems reasonably safe to assume that most of the zinc has not been bio-available in Aquarium AA.

The same reasoning holds for M. Digitata grown in aquarium BB. This aquarium has an extremely high zinc (but also copper) content caused by the use of tap water and old galvanized plumbing in the house. It has nothing to do with a salt mix because only natural seawater was used.

Besides that, if the zinc and copper in that aquarium would have been bio- available then all the corals would have been severely stressed and death would have been a likely scenario because a combination of copper and zinc in a very low concentration elicit synergistic effects of sub lethal toxicity on the zooxanthellae [31]. This has not happened and this fact alone is sufficient to conclude that these metals in this system are also almost not bio-available.

The apparent elevated zinc concentration in A. Latistella grown in Aquarium CC and M. Digitata grown in the Waikiki system despite the low zinc in water concentration of < 1 ppb already indicate that some of the zinc concentrations in skeletons were skewed towards a higher concentration and possible explanations for this has already been given above.

So basing on the data for zinc it seems very likely that most of the zinc has not been bio-available for the examined species in the above aquaria.


The concept of speciation, bio-availability and toxicity has been introduced and is far from complete. Also far from complete is the data presented here, but could be the topic of another article.

Nevertheless within the space of this article and its framework it has been shown that toxicity and bio-availability depend strongly on speciation. If speciation is such that it is not bio-available then it might not be available for some of the biological processes requiring such an element.

The analyses clearly show that copper is far from bio-available and can not exert directly its toxic effect. The same is very likely to be true for zinc: Especially in the case of aquarium BB where there was a very high zinc and copper concentration, sufficient to kill or cause at least clearly visible severe stress to corals if the metals would have been bio-available. Again, it was demonstrated by that aquarist that it was not bio-available – the animals did not die. This is perhaps a far more important observation then the analyses of the coral fragments which only confirm the observations.

Elevated concentration of metals, which might be highly toxic if they were bio-available, are very common in closed systems such as aquariums which we keep. This is true even in systems using natural seawater such as aquaria AA and BB.

Most hobbyists have demonstrated by keeping corals and other invertebrates alive that these metals at elevated levels are not always bio-available and appear not to pose a direct threat to our invertebrates.

The non bio-available speciation of these metals was either already present when the salt mix was dissolved e.g. by being precipitated or became so later by the presence of ligands such as humic acids, carbohydrates and proteins in the water column, and if that had not been enough then organisms such as algae and bacteria might have transformed them further into non bio-available forms. Also, organic and inorganic particles might have rendered the metals non bio- available by adsorbing them on their surface.

Analyses of fragments can confirm this non bio-availability and can be an important tool for those who want to verify this.

The above presented data highly suggests that our aquaria are not always toxic due to elevated levels of heavy metals such as zinc and copper simply because they are not always bio-available enough. The data even suggests that there is a highly reduced bio-availability of at least some of the essential heavy metals when compared to natural seawater.

Now if we go back to the title of this article: Toxicity of trace elements: Truth or Myth?

I have to conclude so far that with respect to copper and zinc it appears to be, for systems and corals comparable to the ones used for this study, more a myth than a truth. It could be that further studies might show it to be the other way round, but so far these are the only data published for organisms grown during a prolonged period in water which is very representative with respect to heavy metals for many aquaria all over the world.

In some aquariums, corals do not grow well due to elevated concentrations of phosphate or other like substances, but not certain heavy metals like copper and zinc when they are not fully bio-available.


  1. The Composition Of Several Synthetic Seawater Mixes by Marlin Atkinson and Craig Bingman Aquarium Frontiers, March 1999.
  2. Unpublished results obtained for internal use by Salifert 1993 – 2002
  3. It’s (In) The Water by Shimek, R.L., Reefkeeping February 2002.
  4. Our Coral Reef Aquaria – Our Own Personal Experiments in the Effects of Trace Element Toxicity by Shimek, R.L., Reefkeeping August 2002.
  5. Chemistry, Toxicity, and Bioavailability of Copper and Its Relationship to Regulation in the Marine Environment P.F Seligman, ed. and A Zirino, ed. November 1998
  6. Copper Binding Abilityof Suwannee River Humic Acid in Seawater Kogut, M.B., MIT 2000.
  7. Cu speciation in estuaries and coastal waters Kogut, M.B., Thesis MIT 2002.
  8. 1999 Progress Report: Biogeochemical Control of Heavy Metal Speciation and Bioavailability in Contaminated Marine Sediments Shine, J.P., EPA Grant # R825220, December 2, 1996–December 1, 2001
  9. Measuring and modeling zinc and cadmium binding by humic acid., Oste L.A., Temminghoff E.J., Lexmond T.M., Van Riemsdijk W.H., Department of Environmental Sciences, Wageningen University, The Netherlands. Anal Chem 2002 Feb 15;74(4):856-62.
  10. Effect of copper on algal communities from oligotrophic calcareous streams. Guasch H., Maria P. and Sabater, S. University of Barcelona, Spain. J. Phycol. 38, 241-248 (2002).
  11. Copper speciation and toxicity in Macquarie Harbour, Tasmania: an investigation using a copper ion selective electrode, Erikson, R.S. ,, Australia. Marine Chemistry, Vol. 74 (2-3) (2001) pp. 99-113.
  12. Acquisition and Utilization of Transition Metal Ions by Marine Organisms, Butler, A. , Science 281, 207 1998.
  13. Metal Speciation Studies in a Brackish / Marine Interface System, SCOULLOS, M.J., PAVLIDOU, A.S., University of Athens, Global Nest: the Int. J. Vol 2, No 3, pp 255-264, 2000.
  14. The utility of the terms “bioavailability” and “bioavailable fraction” for metals, Meyer, J.S., Marine Environmental Research , Vol. 53 (4) (2002) pp. 417-423, Short communication.
  15. Development of Biological Criteria for Coral Reef Ecosystem Assessment, Bioaccumulation of metals, phosphorus. Jameson, S.C., et al., For EPA, Date unknown but >= 2000. Note: Also see Appendix 1 for the references used for this report.
  16. Porites corals as recorders of mining and environmental impacts: Misima Island, Papua New Guinea. Fallon, S.J., J.C. White & M.T. McCulloch, Geochimica Cosmochimica Acta, 66: 45-62 (2002).
  17. A preliminary publication, and different from the above publication on the web: Porites corals as environmental recorders of mining activities on Misima Island, PNG
  18. High resolution analysis and annual variation of trace elements in species of the coral Porites from Mauritius island (Indian Ocean). Publ. Serv. Geol. Lux., Vol. XXIX Proc. 2nd Europe Regional Meeting, ISRS, pp.129-140 (1995) by Immenhauser-Potthast, I.
  19. Metal content on the reef coral Porites astreoides: an evaluation of river influence and 35 years of chronology. Bastidas, C. & E. Garcia. Marine Pollution Bulletin, 38: 899.907 (1999)
  20. Heavy metals in corals from Heron Island and Darwin Harbour, Australia. Esslemont, G. Marine Pollution Bulletin, 38: 1051-1054 (1999)
  21. Harbour dredging in the Townsville region: cross-evaluation between proxy records in coral skeletons and environmental monitoring records. By Esslemont, G. 2000.
  22. Heavy metals in seawater, marine sediments and corals from the Townsville section, Great Barrier Reef Marine Park, Queensland. Esslemont, G., Marine Chemistry, Vol. 71 (3-4) (2000) pp. 215-231.
  23. Chronology of lead pollution contained in banded coral skeletons. Dodge, R.E. & T.R. Gilbert, Mar. Biol., 82: 9-13 (1984).
  24. Tracing a Mine Tailings Spill Using Heavy Metal Concentrations in Coral Growth Bands: Preliminary Results and Interpretation. Coral Reef Symposium Proceedings, Bali, Indonesia. By David, C.P., 2000.
  25. Heavy metal concentrations in growth bands of corals: a record of mine tailings input through time (Marinduque Island, Philippines). David, C.P., Marine Pollution Bulletin (2003), 46(2), 187-196.
  26. Anti-Caking Agents in Salt
  27. Anti-caking Treatment of Salt
  28. Effects of major nutrient additions on metal uptake in phytoplankton. Wen-Xiong Wang and R.C.H. Dei, Environmental Pollution Volume 111, Issue 2 , 2001, Pages 233-240.
  29. Ionic strength effects in biosorption of metals by marine algae. Schiewer , S. and M. H. Wong , Chemosphere Volume 41, Issues 1-2 , 2000, Pages 271-282.
  30. Necessary Nutrition, Foods and Supplements, A Preliminary Investigation. Shimek, R.L. Aquarium Frontiers 2001.
  31. Effects of the Heavy Metals Copper and Zinc on Zooxanthellae Cells in Culture. Goh, B.P.L., and L.M. Chou, Environmental Monitoring and Assessment 44 (1-3): 11-19, February 1997


Submit a Comment

Your email address will not be published. Required fields are marked *