Marine pure ceramic biomedia

[sowildpaul]

Well, after I read the whole thread, I would say ceramic stuffs such as rocks and biomedia are 100% safe after all. Thank God.

[rogergolf66]

Originally Posted by Maximus

I just received a response from Paul at Cermedia (makers of Marinepure). He gave me permission to post his email:

Thank you for your question. I have been following several threads discussing aluminum and it is all very interesting. Our product does contain Alumina, which is Aluminum tied up with oxygen in ceramic form, not the metallic form of Aluminum, or aluminum ion as measured by the Triton test. Most ceramics will contain some alumina. Our products have been used many years with great success by sophisticated reefers and they have never seen a problem.

I have run an 8x8 block for years I will post here what my triton results r when I get them back. My system is simple no GFO or carbon running so no other places the aluminum can come from if I have high levels

[Randy Holmes-Farley]

Originally Posted by sowildpaul

Well, after I read the whole thread, I would say ceramic stuffs such as rocks and biomedia are 100% safe after all. Thank God.

What do you base that on?

[Randy Holmes-Farley]

Originally Posted by Maximus

I just received a response from Paul at Cermedia (makers of Marinepure). He gave me permission to post his email:

Thank you for your question. I have been following several threads discussing aluminum and it is all very interesting. Our product does contain Alumina, which is Aluminum tied up with oxygen in ceramic form, not the metallic form of Aluminum, or aluminum ion as measured by the Triton test. Most ceramics will contain some alumina. Our products have been used many years with great success by sophisticated reefers and they have never seen a problem.

As I stated in the other thread...

OK, thanks for posting that. So one would expect that it will release aluminum ions in a similar way that Phosguard and other aluminum oxide products do.

That doesn't mean that it is necessarily a problem, only that it may be the cause of elevated aluminum that some folks using this material observe.

[shermanator]

ICP cannot distinguish between alumina and aluminum: it's not a useful measurement.

And to be clear: no one is suggesting that alumina is being reduced to aluminum in a reef tank, right?

[Randy Holmes-Farley]

No, there is no concern that aluminum metal is around in reef water tested by ICP and/or released by most products. The distinction between aluminum ions and aluminum oxide is not really appropriate as aluminum oxide is entirely aluminum and oxygen ions.

They are using ICP-OES (Optical Emission Spectroscopy; or sometimes called AES (atomic emission spectroscopy).

So the water is processed a little bit to remove particulates and bacteria and such (Ehsan called it a cyclone, I have no idea on the details or how well that works; I filtered through 0.2 micron filter disks when I've used this method (assuming I wanted it filtered), but they do not, apparently), and then it is heated in the plasma to tremendously high temperatures.

Every element is broken into individual ions and it makes no difference what form it really comes in as. Everything is burned down to ions in the plasma and it is the optical emission of the elements that is then detected, with each element radiating specific bands that can then be counted.

[shermanator]

So why would you (or anyone) worry about elevated levels of alumina? It's chemically inert.

[shermanator]

I struggle to see how ICP (Triton) could be useful whatsoever. For example, Randy: don't you dose sodium silicate? How could you know what your silicate (versus silicone dioxide) level is and how to respond/react to the results?

Bottom line: I would not be concerned whatsoever about the potential correlation of elevated aluminum measurements and ceramic medias.

[Randy Holmes-Farley]

It isn't chemically inert. It dissolves to release soluble aluminum ions. The solubility in freshwater is known to be about 2.5 ppm aluminum at pH 8.2, and soluble aluminum ions has clear toxicity to marine organisms in the literature.

I've shown the release of ions experimentally, and people have for many years seen occasional problems with leather corals exposed to whatever is released from aluminum oxide media.

I show the experiments I've done here:

Aluminum and aluminum-based phosphate binders
http://www.advancedaquarist.com/2003/7/chemistry

and, at Seachem's request, I redid some of the experiments with a smaller amount of phosguard and got the same results here:

New Aluminum Release Data for Phosguard
http://archive.reefcentral.com/forum...light=aluminum

[Randy Holmes-Farley]

Originally Posted by shermanator

I struggle to see how ICP (Triton) could be useful whatsoever. For example, Randy: don't you dose sodium silicate? How could you know what your silicate (versus silicone dioxide) level is and how to respond/react to the results?
.

Not sure what you mean by the silicate level. It is easily tested by kit. Hach makes a good silicate kit, for example. I dose it, but not higher silicate, typically, than is present in seawater. Silicate is rapidly depleted from aquaria and sponges and diatoms consume it.

[shermanator]

I should re-state: it's my opinion that alumina should be chemically inert in a reef tank. That I know of, the mechanisms for taking alumina into a soluble aluminum hydroxide either involves strong acid, strong base, or a catalyst.

I'm curious what mechanism your propose and how you can assure ICP results are not from sub-0.2 µm alumina particles. I'll check out your articles...

[shermanator]

Originally Posted by Randy Holmes-Farley

Not sure what you mean by the silicate level. It is easily tested by kit. Hach makes a good silicate kit, for example. I dose it, but not higher silicate, typically, than is present in seawater. Silicate is rapidly depleted from aquaria and sponges and diatoms consume it.

Right, but I'm saying the Triton method is useless for Si (and other atoms, like Al).

[shermanator]

Originally Posted by Randy Holmes-Farley

I show the experiments I've done here:

Aluminum and aluminum-based phosphate binders
http://www.advancedaquarist.com/2003/7/chemistry

How are you determining something is soluble and in solution? Do you define soluble as something that passes through a 0.45 µm filter? In this instance, I would suggest that solubility should mean a dissolution of ions, which I don't see evidence for.

I'm still not sure how you are discounting that the aluminum from the ICP is simply alumina that is < 0.45 µm?

And do you have standard deviations on your measurements (you did mention n=3)? Is 0.05 statistically different from 0.16? What's the rho value?

Sorry if I'm coming across like a jerk -- maybe I'm missing something, but ICP is a pretty lousy method to look at aluminum species. And aluminum oxide versus aluminum hydroxide is an important distinction (since one is known to be toxic).

EDIT - There are methods (fluorimetric) to detect aluminum ions, have you tried any of these samples in such a test?

[Randy Holmes-Farley]

Originally Posted by shermanator

Right, but I'm saying the Triton method is useless for Si (and other atoms, like Al).

Why would it be useless. In the article above, I show how I use the exact same method (ICP-AES) for aluminum, including showing the exact peaks I observed against a background of seawater without apparent aluminum. I based the quantitation on standards that I ran on the same machine at the same time and showed them:




Figure 2: The ICP emission spectrum between 237.2 and 237.4 nm of a sample from my aquarium (blue) and the same sample after being spiked with 0.5 ppm aluminum (red).



Figure 2 shows some typical ICP data for these samples. The blue line represents the emission spectrum displayed by my aquarium water sample. It is actually the average of three testing replicates taken from a single water sample. The red line represents the same sample with 0.5 ppm aluminum spiked into it. Clearly, the spiked aluminum gives a clearly defined signal exactly where it is supposed to be, while the aquarium water sample shows no such peak. This same conclusion applied at each wavelength examined, and for the Instant Ocean samples as well. From these data, I conclude that the concentrations of aluminum in my aquarium sample and in Instant Ocean are less than 0.05 ppm aluminum.

Nevertheless, the device software for the ICP is what most laboratories will use to quantify ions in high volume samples submitted for analysis. When I used the Varian software to determine the concentration in these samples, it integrated the noise in these samples to the equivalent of 0.05 – 0.12 ppm aluminum for both samples (and for each of several wavelengths used). Likewise, 18 mega-ohm deionized water (very pure) came out at 0.04 to 0.11 ppm aluminum. Consequently, the background noise, and the reality or not of the emissions being quantified must be carefully considered when reporting values near the noise limit for the device.

Was this an important issue for the samples analyzed for aluminum by Shimek? Is that why the samples all looked fairly similar in terms of aluminum concentration? Do any of those reported values represent real determinations, or simply background noise? I don’t know, but I am skeptical. Since I am not aware of any other measurements of aluminum in reef aquaria, I believe that we are left without knowing what the concentrations are, except that in the case of my aquarium at least, the concentration is ≤ 0.05 ppm.

[Randy Holmes-Farley]

Originally Posted by shermanator

How are you determining something is soluble and in solution? Do you define soluble as something that passes through a 0.45 µm filter? In this instance, I would suggest that solubility should mean a dissolution of ions, which I don't see evidence for.

I'm still not sure how you are discounting that the aluminum from the ICP is simply alumina that is < 0.45 µm?

And do you have standard deviations on your measurements (you did mention n=3)? Is 0.05 statistically different from 0.16? What's the rho value?

Sorry if I'm coming across like a jerk -- maybe I'm missing something, but ICP is a pretty lousy method to look at aluminum species. And aluminum oxide versus aluminum hydroxide is an important distinction (since one is known to be toxic).

EDIT - There are methods (fluorimetric) to detect aluminum ions, have you tried any of these samples in such a test?

I did not try to prove the aluminum ions were individual, and they may not be. Almost certainly they will quickly be at least partially bound to organics, if they are not already. They may be fine aluminum oxide particulates, or even aluminum trihydroxide and tetrahydroxide individually.

In normal seawater, aluminum is expected to take the forms Al(OH)4- and Al(OH)3, according the Millero in "Chemical Oceanography". I'm not sure why you are certain it is different here.

[shermanator]

Originally Posted by Randy Holmes-Farley

Why would it be useless. In the article above, I show how I use the exact same method (ICP-AES) for aluminum, including showing the exact peaks I observed against a background of seawater without apparent aluminum.

My concern is that this analysis implies that anything that any Al or Si species that passes through a 0.45 µm filter is ionic. I don't agree with that premise. If you (or someone) took a sample of alumina in 18-ohm water, filtered it and ran an ICP, you'd see Al show up (at least in my opinion, because alumina has fines < 0.45 µm) ). Would you suggest that is ionic aluminum?

Another experiment would be to do the alumina filtration as described and see if it poses any issues to reef life. To my knowledge, you and others have only shown that ionic aluminum is problematic.

[Randy Holmes-Farley]

FWIW, Triton claims the Limit of Detection for their method for aluminum is about 100 ng/kg, or about 0.1 ppb.

That is a bit above the NSW value, but it is far, far below the levels that some folks have gotten of >0.1 ppm (>100 ppb). So I don't really think that the limit of detection or quantitation for is an issue for aluminum levels above 0.1 ppm.

It wasn't in my case, and I accept that it isn't for them (although I can only rely on their assertion of their LOD).

[shermanator]

Originally Posted by Randy Holmes-Farley

I did not try to prove the aluminum ions were individual, and they may not be. Almost certainly they will quickly be at least partially bound to organics, if they are not already. They may be fine aluminum oxide particulates, or even aluminum trihydroxide and tetrahydroxide individually.

Right, this is exactly my point. There is a disconnect.

It's known that ionic aluminum is toxic. You show Phosguard leads to some sort of atomic aluminum show up in the ICP. It could simply be small particle alumina (which carries no risk that I know of).

In normal seawater, aluminum is expected to take the forms Al(OH)4- and Al(OH)3, according the Millero in "Chemical Oceanography". I'm not sure why you are certain it is different here.

I'm not suggesting anything different. I'm arguing that tiny particles of aluminium oxide (alumina, Al2O3) that are < 0.45 µm is not (and will not under normal reef tank conditions lead to) ionic aluminium species.

[Randy Holmes-Farley]

Originally Posted by shermanator

My concern is that this analysis implies that anything that any Al or Si species that passes through a 0.45 µm filter is ionic. I don't agree with that premise. If you (or someone) took a sample of alumina in 18-ohm water, filtered it and ran an ICP, you'd see Al show up (at least in my opinion, because alumina has fines < 0.45 µm) ). Would you suggest that is ionic aluminum?

Another experiment would be to do the alumina filtration as described and see if it poses any issues to reef life. To my knowledge, you and others have only shown that ionic aluminum is problematic.

So lets' back up. I'm not making any claims about the form of the aluminum

here's the history:

1. For many years, people had been noticing the leather corals sometimes closed up when using Phosguard or other aluminum oxide phosphate binders. That predates me. That demonstrates that there is at least sometime a reason to be concerned, regardless of any form issues or measurements. It was a hobbyists result that was fairly well recognized at the time (1980's - 1990's)

2. I thought that one possible explanation for the issue was released aluminum, since there is known solubility of aluminum in fresh water and seawater. But that was just a hypothesis.

3. I measured to see if there was any aluminum that came off of phosguard in any form. There was, and much of it was not filterable at 0.45 microns. Still don't know the form, but it is not particulates by the standard definition. It may be colloids.

4. I added an amount of soluble aluminum to a test tank. I added amounts which were below the known solubility of aluminum in fresh water at the same pH. I added it sowly, from lower to higher concentration. At the highest concentration (0.5 ppm) I observed the same phenomenon that reefers observed: leather coral closed up, etc.

5. Since it was happening at about the same concentration of aluminum observed with the Phosguard, I concluded that the aluminum may have been the explanation of what was happening when people used Phosguard.

Obviously it does not prove exact cause and effect, but it links the entire hypothesis together, and is enough, IMO, to be a caution.

However, I am aware that trace metal toxicity is very dependent on the form the metal takes and on organic binding of the metal.

[Randy Holmes-Farley]

Originally Posted by shermanator

Right, this is exactly my point. There is a disconnect.

It's known that ionic aluminum is toxic. You show Phosguard leads to some sort of atomic aluminum show up in the ICP. It could simply be small particle alumina (which carries no risk that I know of).

I'm not suggesting anything different. I'm arguing that tiny particles of aluminium oxide (alumina, Al2O3) that are < 0.45 µm is not (and will not under normal reef tank conditions lead to) ionic aluminium species.

Well, that may be. They may be toxic as very fine particles. I do not believe I ever claimed otherwise anywhere, although I think it quite likely that there is soluble aluminum present since it is well known in other contexts (like the fresh water solubility) and soluble aluminum gave the same effects at similar concentrations. Why do you discount that?

Do you have a different hypothesis to explain the tox data that has been known for a very long time that sometimes followed Phosguard use?

[shermanator]

Originally Posted by Randy Holmes-Farley

4. I added an amount of soluble aluminum to a test tank.

I assume you started with an Al(III) source of some sort? If you (or anyone) did this starting with Al2O3 and saw a problem with any reef life, I would take back everything I've written. Well, not everything, but I would agree that alumina-based products are of concern.

Again, sorry if I came across badly. From a chemistry perspective, I have trouble with the idea that Al2O3 will (again, in a reef tank) ever wind up as ionic Al(III).

[shermanator]

Originally Posted by Randy Holmes-Farley

Do you have a different hypothesis to explain the tox data that has been known for a very long time that sometimes followed Phosguard use?

No, but correlation does not imply causation. [Sorry, couldn't resist.]

I'm fairly new to this hobby (but not new to chemistry), but I know there are many variables to worry about whenever trying to pin down a culprit for anecdotal (non-controlled) reports.

[Randy Holmes-Farley]

Originally Posted by shermanator

I assume you started with an Al(III) source of some sort? If you (or anyone) did this starting with Al2O3 and saw a problem with any reef life, I would take back everything I've written. Well, not everything, but I would agree that alumina-based products are of concern.

Again, sorry if I came across badly. From a chemistry perspective, I have trouble with the idea that Al2O3 will (again, in a reef tank) ever wind up as ionic Al(III).

But that is exactly what people did, or at least, they started with Phosguard (not sure of its exact crystal structure, and certainly not sure that 100.000000% of it is corundum and not oxyhydroxides on the surfaces, which are far more soluble), and they saw a problem.

What exactly caused that problem is, in some ways, immaterial.

[shermanator]

Originally Posted by Randy Holmes-Farley

But that is exactly what people did, or at least, they started with Phosguard (not sure of its exact crystal structure, and certainly not sure that 100.000000% of it is corundum and not oxyhydroxides on the surfaces, which are far more soluble), and they saw a problem.

What exactly caused that problem is, in some ways, immaterial.

Someone setup a control tank (like you did in your article) and showed that the filtrate of Phosguard causes problems?

In your article you have a nice, controlled experiment (but I suspect you used Al(III)). When someone posts that they added Phosguard and all hell broke loose, that of course isn't a controlled experiment.

Side note: Advanced Aquarist needs a methods section.

[Randy Holmes-Farley]

Originally Posted by shermanator

No, but correlation does not imply causation. [Sorry, couldn't resist.]

I'm fairly new to this hobby (but not new to chemistry), but I know there are many variables to worry about whenever trying to pin down a culprit for anecdotal (non-controlled) reports.

I'm surprised you have such an issue with it.

This paper shows that the solubility of corundum (least soluble of all the aluminum oxide hydroxides) at pH 4 is about 10^-3 M or 27 ppm.

Are you really so confident that at pH 8.2, the 0.4 ppm that I observed cannot be dissolved? Pankow ("Aquatic Chemistry Concepts") shows a solubility graph vs pH for all aluminum species, and pH 8.2 the total is lower than at pH 4, but by less than a factor of 100. So I just don't see how 0.4 ppm is out of the question, especially since the surfaces of Phosguard may well be hydrous oxide/hydroxides and not pure corundum (which are more soluble than corundum by far).

http://www.clays.org/journal/archive...6/36-5-391.pdf