DIY sulphur denitrator

[robert]

I'm not sure what the cloudiness was, but bacteria is a reasonable guess, alive or dead.

FWIW, bacteria in the water are generally desirable as food, unless somehow you drive the growth of a pathogenic species (which has, perhaps, happened in a few isolated cases with organic carbon dosing).

I have no reason to doubt that at least some corals ingest at least some bacteria. I do however doubt the significance of water born bacteria to the diet of most of the corals we keep. Bacterial counts in pristine overlying reef water are very small and I would guess insufficient to the metabolic needs of the coral. That fact that some water born bacterial strains can be detrimental to corals is without question.

If bacteria were a generally desirable food source, you would think corals would flourish in bays and estuaries. Some can - most can't.

This fundamentally is the reason I prefer the sulphur approach. Stimulating water column bacterial growth without knowing what the bacteria are is a game of chance. I am sure the dangers might be offset somewhat by bacterial dosing. Introducing a know bacteria that is not pathalogic. But again your assuming a more dominant strain is not already present in the system.

In my systems I strive to keep nutrients high and water column bacterial load low.

 

[robert]

I found this to be interesting...

http://www.captivereefs.com/forum/f...bacteria-swim-coral-mucus-21766/#.Vdvul_lViko

 

[robert]

On a final note - the sulphur denitrator is not totally sequestering in the aspect that it does convert substantial quantites of nitrate by exporting nitrogen gas which is blown off and sulphate which precipitates in the body of the denitrator, and in the case if the mixed bed reactor MAP - struvite (magnesium ammonium phosphate) near the contact between the sulphur and coral chunks.

The addition of GFO or iron furthure sequesters phosphate in an ph environment much more favorable to phosphate binding.

Really not at all like carbon dosing.

 

[Oceansize]

Robert I'm glad you added that last comment because otherwise I would have thought I was going crazy. The entire reason I became interested in sulphur denitrators was due to the conversion of nitrates to nitrogen gas and the resultant _export_ of that gas, not sequestration.

I was intrigued by the notion that the entire nitrogen cycle beginning with ammonia and normally ending with nitrates and water changes could instead end with a gas effortlessly floating out of my tank (and far fewer water changes). Pretty amazing that: a fish poops in my tank and after a few hours it ultimately converts to a gas that EXPORTS ITSELF from my tank by just floating away on air. Brilliant.

So thank you for the clarification! LOL

I've also noticed now that my denitrator is really broken in well that my skimmer barely produces any skimmate, and what skimmate there is, while remaining ugly green or brownish, doesn't really have a nasty smell at all. Interesting that a filter that is supposed to address the _final_ part of the nitrogen cycle (converting nitrates to nitrogen) helps ease the load on a filter that _precedes_ the nitrogen cycle (removing proteins before they become ammonia). I have zero chemistry background so if you have an explanation for why this is, I'd love to hear it.

I'm also curious what signs you look for to tell you that your chamber is ready for a flush.

 

[Randy Holmes-Farley]

I have no reason to doubt that at least some corals ingest at least some bacteria. I do however doubt the significance of water born bacteria to the diet of most of the corals we keep. Bacterial counts in pristine overlying reef water are very small and I would guess insufficient to the metabolic needs of the coral. That fact that some water born bacterial strains can be detrimental to corals is without question.

If bacteria were a generally desirable food source, you would think corals would flourish in bays and estuaries. Some can - most can't.
.

I don't claim expertise in this area, but others who do seem to think differently:

http://reefkeeping.com/issues/2003-01/eb/index.htm

from it:

"Given the importance of bacteria as a food source in marine ecosystems, it might not be surprising to learn that they are also a primary food source for corals. It has been found that bacteria alone can supply up to 100% of both the daily carbon and nitrogen requirements of corals. All corals studied consume dissolved organic material, bacteria, and detrital material. This is more than can be said for any other food source, including zooplankton and light."

 

[Randy Holmes-Farley]

On a final note - the sulphur denitrator is not totally sequestering in the aspect that it does convert substantial quantites of nitrate by exporting nitrogen gas which is blown off and sulphate which precipitates in the body of the denitrator, and in the case if the mixed bed reactor MAP - struvite (magnesium ammonium phosphate) near the contact between the sulphur and coral chunks.

The addition of GFO or iron furthure sequesters phosphate in an ph environment much more favorable to phosphate binding.

Really not at all like carbon dosing.

First, sulfate doesn't precipitate in the reactor. It is the third most abundant ion in seawater by mass and is extremely soluble.

I've not seen any evidence that MAP forms in denitrators, but it might. Why do you think it present?

Not sure what you mean by a favorable environment for GFO binding. You mean lowish pH in the denitrator? Why do you think that favorable for phosphate to bind to GFO?


Obviously much of the nitrate gos to N2 gas, as nitrate does when anaerobic bacteria consume organic carbon in a low O2 environment such as rock pores or sand:

organic + 124 NO3- + 124 H+ → 122 CO2 + 70 N2 + 208 H2O


So as the bacteria consume the sulfur in a denitrator, convert nitrate to N2, and release sulfate, what else do you suppose they are doing? They are growing in mass. That's about all bacteria can do with energy. Its not like the work out or think really hard. They grow. And as they grow they will be sloughed off into the water. There's no other way it can be. I've never actually heard someone claim they are not released.

 

[robert]

Not sure what you mean by a favorable environment for GFO binding. You mean lowish pH in the denitrator? Why do you think that favorable for phosphate to bind to GFO?
.

Binding of phosphate, silicate, and fluoride on ferric oxide hydroxide (GFO) Species shown are surface species.
(6.0 g FeOOH/L, PT= 10-3 M, SiT = 8×10-4 M.) Sigg and Stumm, 1981

Phosphate Adsorption by GFO
pH—— % bound (%efficiency)
4 ——- ~93%
5 ——- ~82%
6 ——- ~62%
7 ——- ~41%
8 ——- ~19%
8.8 —— ~0 %

GFO is roughly twice as effective at pH 7, compared to pH 8.0, and as pH goes up from 8.0, performance drop off drastically.

 

[robert]

First, sulfate doesn't precipitate in the reactor. It is the third most abundant ion in seawater by mass and is extremely soluble.

"Another by-product of sulfur-based denitrification is sulfate. Passing the water through calcareous gravel precipitates this sulfate and prevents it from entering the aquarium. The sulfur beads can last a very long time, but the chamber of coral gravel will quickly turn to sediment and should be replenished on a regular basis. The use of a fluidized bed design for both the sulfur and the coral gravel reactors should help to reduce the build-up of sediments."

Yes but inside the sulphur denitrator there is an abundance of CA+ ions which bind to the sulphate. Calcium sulphate has poor soluabilty, gypsum precipitation.

See also Calcium Sulfate Precipitation in Biotrickling Filters Treating Hydrogen Sulfide
Loughery, Scott 2012

 

[robert]

I've not seen any evidence that MAP forms in denitrators, but it might. Why do you think it present?

There is some contention on this issue.

Phosphorus removal in a sulfur–limestone autotrophic denitrification (SLAD) biofilter
Ruihua Li, Yulin Yuan, Xinmin Zhan, Bo Liu

"The optimal phosphorus removal in the SLAD biofilter was around 60 %. For the synthetic wastewater containing a PO43−-P concentration of 15 mg L−1, the main mechanism of phosphorus removal was the formation of calcium phosphate precipitates."

However others, because of the increased P removal in the mixed media packed bed reactors suggested MAP deposited at the junction of the mixed sulphur and coral - hence the mixed media design of my DIY filter. As I remember, the presence of MAP was determined by crystal-xray-spectrometry. Struvite (MAP) is a common percipitate in waste-water treatment and is not soluable in our systems.

 

[robert]

Obviously much of the nitrate gos to N2 gas, as nitrate does when anaerobic bacteria consume organic carbon in a low O2 environment such as rock pores or sand:

organic + 124 NO3- + 124 H+ → 122 CO2 + 70 N2 + 208 H2O


So as the bacteria consume the sulfur in a denitrator, convert nitrate to N2, and release sulfate, what else do you suppose they are doing? They are growing in mass. That's about all bacteria can do with energy. Its not like the work out or think really hard. They grow. And as they grow they will be sloughed off into the water. There's no other way it can be. I've never actually heard someone claim they are not released.

If you misinterpreted what I said. Please go back and re-read my post #3. I said "It doesn't add a bacterial load to the water column like carbon dosing does."

I further clarified this in post #10 "The vast majority of bacteria in our systems is in attached bio-films. This is the case in the sulphur denitrator. The bacteria is growing on the media - not in the water column. The flow is slow, meaning very little of the bacterial film is ever dislodged."

I stand by this.

 

[robert]

I don't claim expertise in this area, but others who do seem to think differently:

http://reefkeeping.com/issues/2003-01/eb/index.htm

from it:

"Given the importance of bacteria as a food source in marine ecosystems, it might not be surprising to learn that they are also a primary food source for corals. It has been found that bacteria alone can supply up to 100% of both the daily carbon and nitrogen requirements of corals. All corals studied consume dissolved organic material, bacteria, and detrital material. This is more than can be said for any other food source, including zooplankton and light."

"Detrital food chains are found to predominate in most marine ecosystems, and it has been found that the bulk of the diet of herbivores, as well as their nutritional requirements, comes not from direct consumption of phytoplankton but by the consumption of the adherent periphyton and detrital material that is enriched by attached microbial communities."

The detrital food chain - adherent periphyton and detrital material that is enriched by attached microbial communities.

I think the author is saying dissolved organic material, bacteria, and detrital material is a greater food source than zooplankton and light - I don't think he is saying bacteria alone. Altought it is written that way, I think his intent was obvious.

My point was inarticulately made. My point was "Stimulating water column bacterial growth without knowing what the bacteria are is a game of chance. I am sure the dangers might be offset somewhat by bacterial dosing. Introducing a know bacteria that is not pathalogic. But again your assuming a more dominant strain is not already present in the system." and that in a pristine reef are free bacterial counts are very small and I would guess insufficient to the metabolic needs of the coral.

Please read....http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2253183/

Microbial Ecology of Four Coral Atolls in the Northern Line Islands
Elizabeth A. Dinsdale,#1,2,*Olga Pantos,#1,¤Steven Smriga,3Robert A. Edwards,4,5Florent Angly,1Linda Wegley,1Mark Hatay,1Dana Hall,1Elysa Brown,1Matthew Haynes,1Lutz Krause,6Enric Sala,3Stuart A. Sandin,3Rebecca Vega Thurber,1Bette L. Willis,7Farooq Azam,3Nancy Knowlton,3 and Forest Rohwer1,4,*

The important take away - elevated bacterial count (along with N/P and reduced DOC (excess carbon consumption) was assiciated with decreased reef health, with a shift in the type of bacteria seen over the reef.

 

[Randy Holmes-Farley]

Binding of phosphate, silicate, and fluoride on ferric oxide hydroxide (GFO) Species shown are surface species.
(6.0 g FeOOH/L, PT= 10-3 M, SiT = 8×10-4 M.) Sigg and Stumm, 1981

Phosphate Adsorption by GFO
pH—— % bound (%efficiency)
4 ——- ~93%
5 ——- ~82%
6 ——- ~62%
7 ——- ~41%
8 ——- ~19%
8.8 —— ~0 %

GFO is roughly twice as effective at pH 7, compared to pH 8.0, and as pH goes up from 8.0, performance drop off drastically.

I believe that data is in freshwater, not seawater. I've seen similar data in DI water, but I wouldn't necessarily translate it to seawater where the ionic forms of phosphate as a function of pH are quite different. Same for the surface charge on GFO, and there are also specific ion effects (such as calcium boosting phosphate binding on GFO). The binding of phosphate to GFO in seawater at pH 6 may be higher than at pH 8, but I wouldn't assume it is true without seeing some data.

 

[Randy Holmes-Farley]

"Another by-product of sulfur-based denitrification is sulfate. Passing the water through calcareous gravel precipitates this sulfate and prevents it from entering the aquarium. The sulfur beads can last a very long time, but the chamber of coral gravel will quickly turn to sediment and should be replenished on a regular basis. The use of a fluidized bed design for both the sulfur and the coral gravel reactors should help to reduce the build-up of sediments."

Yes but inside the sulphur denitrator there is an abundance of CA+ ions which bind to the sulphate. Calcium sulphate has poor soluabilty, gypsum precipitation.

See also Calcium Sulfate Precipitation in Biotrickling Filters Treating Hydrogen Sulfide
Loughery, Scott 2012

Not sure who write your italicized part, but it is absolutely incorrect. Let's think about these issues quantitatively.

Assume the tank water starts at 35 ppt salinity and 20 ppm nitrate, so that means about:

420 ppm calcium (10.5 mM)
2700 ppm sulfate (28.1 mM)
20 ppm nitrate (0.33 mM)

If we assume the sulfur denitrator removes all of the nitrate and converts it to N2 gas, here's the equation:

2 H2O + 5 S + 6 NO3- → 3 N2 + 5 SO42- + 4 H+

So for every mM of nitrate removed, we add 0.8 mM sulfate and 0.66 mM H+.

Let's assume the best possible case for calcium release, that ALL of that H+ goes into dissolving CaCO3 (which, if it happened, would mean the pH in the reactor is the same as the tank and not significantly lower):

2H+ + CaCO3 --> Ca++ + 2HCO3-

So for every mM of nitrate removed, we add 0.66 x 0.5 = 0.33 mM of calcium.

So now we compare our starting and ending levels of calcium and sulfate:

420 ppm (10.5 mM) calcium ---> 435 ppm (10.88 mM) calcium a rise of 3.6%
2700 ppm (28.1 mM) sulfate ---> 2777 ppm (28.9 mM) sulfate a rise of 2.9%

So I wouldn't characterize this as either high calcium or high sulfate. I expect a great many reef tanks exceed both of these values, and calcium sulfate does not precipitate. In fact, it is known how much you have to evaporate seawater to get calcium sulfate to precipitate, and the concentrations involved are far higher than these vlaues in the reactor.

Again, you cannot extrapolate from fresh water data. Most ionic solids are far more soluble in seawater than in fresh water. Calcium carbonate, for example, is far more soluble in seawater than in fresh water. The Ksp (solubility product constant) is more than 100 times higher in seawater!

Consequently, any claim that calcium sulfate precipitates in the reactor is incorrect.

 

[Randy Holmes-Farley]

If you misinterpreted what I said. Please go back and re-read my post #3. I said "It doesn't add a bacterial load to the water column like carbon dosing does."

I further clarified this in post #10 "The vast majority of bacteria in our systems is in attached bio-films. This is the case in the sulphur denitrator. The bacteria is growing on the media - not in the water column. The flow is slow, meaning very little of the bacterial film is ever dislodged."

I stand by this.

OK, perhaps we just have to agree to disagree. As they multiply they must go somewhere (IMO) and I don't see anywhere to go except out of the reactor, just like bacteria growing on biopellets or those on and in live rock and sand in the main tank driven by organic carbon dosing.

 

[Randy Holmes-Farley]

There is some contention on this issue.

Phosphorus removal in a sulfur–limestone autotrophic denitrification (SLAD) biofilter
Ruihua Li, Yulin Yuan, Xinmin Zhan, Bo Liu

"The optimal phosphorus removal in the SLAD biofilter was around 60 %. For the synthetic wastewater containing a PO43−-P concentration of 15 mg L−1, the main mechanism of phosphorus removal was the formation of calcium phosphate precipitates."

However others, because of the increased P removal in the mixed media packed bed reactors suggested MAP deposited at the junction of the mixed sulphur and coral - hence the mixed media design of my DIY filter. As I remember, the presence of MAP was determined by crystal-xray-spectrometry. Struvite (MAP) is a common percipitate in waste-water treatment and is not soluable in our systems.

I've not seen anything convincing that suggests struvite will form in a sulfur denitrator, or anywhere in a reef tank. All of the data I've seen is for fresh wastewater systems with hugely high phosphate and significant ammonia concentrations. Your quoted article had phosphate at 15 ppm in freshwater.


What is it about a sulfur denitrator that you think makes such precipitation likely?

The Ksp is:

[Mg++] x [NH4+] x [PO4---]

The magnesium and phosphate won't be higher in the denitrator than in the tank. pH may be lower, which will reduce the precipitation by converting PO4--- to HPO4--. Do you think ammonia is elevated? If the pH is lower, there is a tiny increase in NH4+ from NH3 as the pH drops below 8, but it cannot rise more than about 4% because at 8.0, 96% of it is already present as NH4+.

So unless you think you are generating ammonia in the denitrator, I cannot see any reason you'd precipitate more struvite there than anywhere else in the reef tank.

 

[Randy Holmes-Farley]

Struvite (MAP) is a common percipitate in waste-water treatment and is not soluable in our systems.

FWIW, I believe that it is not clear that struvite will be insoluble in our reef systems. It is not a waste water system with high ammonia and phosphate. Only if ammonia is significantly above 0.05 ppm and phosphate above 0.1 ppm will it be insoluble.

Let's look at some real values.

The Ksp for struvite in seawater is 10-(13.08) = 8.3 x 10^(-14), which is from:

Solubility of Struvite in Seawater
http://www.the-conference.com/JConfAbs/5/449.pdf

where they also claim:
"As a result of this investigation and our experimental data about unseeded precipitation of calcium and magnesium phosphates from modified seawater solutions (Golubev et al.,1999), no magnesium (ammonium) phosphate can be inorganically precipitated in modern marine environments..."

But let's use their Ksp and see how our reef tanks match up.

The Ksp is:

[Mg++] x [NH4+] x [PO4---]

Magnesium in seawater is 1280 pppm or 0.053 M

Let's examine phosphate at 0.1 ppm = 0.1 mg/L = 1 x 10^(-6) M at pH 8.1 only 20% of it is present as PO4--- (and that number drops as the pH lowers), so the PO4--- is about 2 x 10^(-7)

Let's also examine ammonia at a fairly high 0.05 ppm, or 2.8 x 10^(-6) M and assume it is all ammonium

So the solubility product in this case is 0.053 x 2 x 10^(-7) x 2.8 x 10^(-6) = 2.9 x 10-14.

So the solubility product is lower than the Ksp in seawater, indicating that it will be soluble (and won't precipitate) under these conditions.

 

[Randy Holmes-Farley]

The important take away - elevated bacterial count (along with N/P and reduced DOC (excess carbon consumption) was assiciated with decreased reef health, with a shift in the type of bacteria seen over the reef.

I don't deny the possibility of boosting a pathogenic bacteria with organic carbon dosing, but I only know of one example where it might have happened.

But I do not agree that reef tanks, even with carbon dosing, have bacteria levels vastly higher than natural reefs.

http://www.advancedaquarist.com/2011/3/aafeature

"Aquaria subjected to active filtration via skimming present water column bacteria populations that are approximately 1/10 of those observed on natural reefs. "

They also found that bacteria went up with carbon dosing, but not all that much, which may be partly due to predation by larger plankton and other organisms.

On the positive side, I have been able to keep sponges thriving and expanding in size, despite the claim that they generally need a special plankton feedings which I do not do.

 

[robert]

I believe that data is in freshwater, not seawater. I've seen similar data in DI water, but I wouldn't necessarily translate it to seawater where the ionic forms of phosphate as a function of pH are quite different. Same for the surface charge on GFO, and there are also specific ion effects (such as calcium boosting phosphate binding on GFO). The binding of phosphate to GFO in seawater at pH 6 may be higher than at pH 8, but I wouldn't assume it is true without seeing some data.

You should reconize this from your paper on Phosphate in the Reef Aquarium.

Which is mirrored in Biogeochemistry by William H, Schlesinger.
In the second diagram, the first diagram is fresh water and the last artificial sea water.
From my post:
Phosphate Adsorption by GFO
pH—— % bound (%efficiency)
4 ——- ~93%
5 ——- ~82%
6 ——- ~62%
7 ——- ~41%
8 ——- ~19%
8.8 —— ~0 %

if this is freshwater data, may we deduce that it GFO binds preferentialy to the H2PO4 species - other wise the binding efficiency would be flat as a function of pH? It is difficult to assess how much of the HPO4 is bound. looking at the data and the curves, it would suggest that GFO doesn't bind HPO4 effectively at all; perhaps a few percent.

For the salt water, The fraction of phosphate as H2PO4 is 10% at a pH 7 of and virtually 0 as you approach 8. If it holds that this binding affinty held and if the lack of binding affinity for HPO4 holds for salt water, then this additional fraction which could be cleared by GFO running in a pH 7 environment could be a very significant multiple of what can be cleared at 8.

I think its worth testing.

 

[robert]

Not sure who write your italicized part, but it is absolutely incorrect. Let's think about these issues quantitatively.

Assume the tank water starts at 35 ppt salinity and 20 ppm nitrate, so that means about:

420 ppm calcium (10.5 mM)
2700 ppm sulfate (28.1 mM)
20 ppm nitrate (0.33 mM)

If we assume the sulfur denitrator removes all of the nitrate and converts it to N2 gas, here's the equation:

2 H2O + 5 S + 6 NO3- → 3 N2 + 5 SO42- + 4 H+

So for every mM of nitrate removed, we add 0.8 mM sulfate and 0.66 mM H+.

Let's assume the best possible case for calcium release, that ALL of that H+ goes into dissolving CaCO3 (which, if it happened, would mean the pH in the reactor is the same as the tank and not significantly lower):

2H+ + CaCO3 --> Ca++ + 2HCO3-

So for every mM of nitrate removed, we add 0.66 x 0.5 = 0.33 mM of calcium.

So now we compare our starting and ending levels of calcium and sulfate:

420 ppm (10.5 mM) calcium ---> 435 ppm (10.88 mM) calcium a rise of 3.6%
2700 ppm (28.1 mM) sulfate ---> 2777 ppm (28.9 mM) sulfate a rise of 2.9%

So I wouldn't characterize this as either high calcium or high sulfate. I expect a great many reef tanks exceed both of these values, and calcium sulfate does not precipitate. In fact, it is known how much you have to evaporate seawater to get calcium sulfate to precipitate, and the concentrations involved are far higher than these vlaues in the reactor.

Again, you cannot extrapolate from fresh water data. Most ionic solids are far more soluble in seawater than in fresh water. Calcium carbonate, for example, is far more soluble in seawater than in fresh water. The Ksp (solubility product constant) is more than 100 times higher in seawater!

Consequently, any claim that calcium sulfate precipitates in the reactor is incorrect.

Your analysis in free water is correct. I don't know the details of how the author derived his conclusion, but it was my conclusion that calcium sulphate precipitated. He doesn't say this.

I can say calcium sulphate is produced in mats and biofilms in the marine anerobic environment and in quantity, despite its soluabilty; And not just through brine, volcanic or evaporative mechanisms. But as you also pint out the increase sulphate as a percentage is minimal and is anecdotally of no consequence.

 

[robert]

Sorry about that last post - the browser postd the wrong quote...