Ozone is often used by reef aquarists to "purify" their tank's water. To most aquarists this means making the water visually clearer, so they elect to start using ozone to improve clarity. Not everyone notices such improvements, however, and some aquarists have been disappointed. What sort of changes can be expected upon initiating ozone?

This article is the third in a series that discusses the details of ozone and its use in reef aquaria:

Ozone and the Reef Aquarium, Part 1: Chemistry and Biochemistry
Ozone and the Reef Aquarium, Part 2: Equipment and Safety
Ozone and the Reef Aquarium, Part 3: Changes in a Reef Aquarium upon Initiating Ozone

The first article in this series detailed what ozone is and how it reacts with seawater. It also related ozone's perceived benefits to the actual chemical and biochemical changes that it can cause. The second article described the equipment and processes used to safely and effectively apply ozone in reef aquaria. This third article details the changes observed in my aquarium system upon initiating ozone. Through pictures and measurements, it is hoped that aquarists can attain an understanding of ozone's likely benefits before spending the time and money necessary to try it for themselves. While the results will not translate directly to every reef aquarium, my husbandry practices are not unusual and these results seem to reflect those found by a wide range of aquarists who have used ozone similarly.

The sections are:

• Introduction
• ORP Changes
• ORP Measurement
• Water Clarity
• Algae Growth
• Response of Organisms
• Oxygen Levels
• Effect of Air Drying
• Disinfection of Water by Ozone Treatment
• Activated Carbon Treatment
• Conclusions About Reaction Chambers
• Details of My Aquarium System
• Summary
• References

Introduction

The following sections detail a variety of visual and chemical changes that I noted in my aquarium upon initiating ozone. In most cases, I compare before and after measurements or photographs to show the effects. Most of the "after" cases were two weeks after initiating ozone, but some were recorded or continuously monitored for up to two months.

In order to evaluate whether the effects obtained in my aquarium will translate to other aquaria, it is important to understand how my aquarium is maintained and how the ozone was used.

Ozone was supplied by a 100 mg/hr Aqua Medic corona discharge ozone generator. In most cases, an Optima air pump sent the air through a Red Sea air dryer (500 g model) and then into the ozone generator. I also experimented with a Tetratec Deep Water air pump, with little difference noticed. The air/ozone mixture was then mixed with aquarium water in three different ways: a Coralife ozone reactor, my skimmer (ETS 800 Gemini), or a tubing reactor of my own design. The effluent air and water were passed over granular activated carbon (GAC) before being returned to the room and the aquarium, respectively (except when using the skimmer, where no GAC was employed).

My aquarium system consists of two main tanks (120 gallons and 90 gallons) and a variety of sump and support systems, including skimming, large macroalgae and sand-filled refugia. The aquaria contain a variety of fish, live rock, sand and invertebrates, including both hard and soft corals. In short, they are typical "mixed" reef aquaria. Near the end of this article is a section that details the nature of my aquarium system so that aquarists can understand whether their system deviates from mine in some important way(s).

ORP Changes

When ozone is first added to reef aquaria, ORP (oxidation reduction potential) rises. What exactly this means on a molecular level is complicated and not well understood, but it is the expected result of oxidizing the various redox active chemical species in the water. Some aquarists see a big rise in ORP - large enough that they need to control the ozone generator so that the ORP does not rise too high (>450-500 mV). Others see a relatively small rise, and still have ORP in the 300 mV range even after using the generally recommended amount of ozone. Some even find that it remains in the upper 200 mV range. This variability is important for aquarists to keep in mind, and they should not conclude that there is a problem if the ORP remains fairly "low."

In the absence of ozone, my aquarium seems to have an ORP that ranges in the mid- to upper 200 mV range. While I monitored ORP carefully when initiating ozone to be sure not to overdo the addition, such concern was unfounded. Even when adding ozone at full production (100 mg/h as claimed by Aqua Medic; equivalent to about 0.5 mg/h/gallon of display volume or 0.3 mg/h/gallon in the entire system), the ORP never rose above about 335 mV. Even with newly dried desiccant in the air dryer, a freshly cleaned skimmer and ozone added to any of the devices that I used (Coralife ozone reactor, straight into a skimmer, or my tubing reactor), the ORP was never higher than this. Even when used through the skimmer with no activated carbon treatment of the effluent, the ORP was no higher. Consequently, I did not need to use the ozone on a controller.

Was ozone being added? Clearly, yes. The water's ORP exiting my tubing reactor measured 680 mV. I could also detect an ozone/OPO residual of 0.1 - 0.24 ppm chlorine equivalents (details of such test methods are provided in air flow section of the previous article). So the ozone was having the desired effect on the water in the reactor.

Would the ORP have risen more with additional ozone? Certainly, yes. But using higher amounts of ozone causes more risk of harm from overdose, and it may not further improve the primary reason for using it: water clarity. More ozone might have resulted in more disinfection of the water in the ozone reaction chamber, and possibly a further reduction in dissolved organic levels, but I am not sure that either of those is necessarily beneficial.

ORP Measurement

Because many aquarists focus on ORP levels when using ozone, a final repeated caution about the inherent complications of ORP measurement is warranted, along with some examples. ORP is not a simple measurement like pH. The metal probe's tip is very sensitive to what is bound to it, and it can take days for it to become coated with organics and other materials in reef aquarium water. So, when the probe is first put into a reef system, the ORP can drift for quite some time as these processes slowly take place. Moreover, every time the probe is calibrated or moved into a different solution, it can again take days to re-establish itself in the aquarium.

For all measurements I used a Pinpoint ORP probe attached to an Orion 710A pH/ORP meter. Proper operation was checked using Pinpoint's 400 mV single-use calibration solutions.

Here are some examples of apparent ORP measurement complications:

1. After using ozone for several weeks in my system, the ORP was running in the 300-330 mv range, depending on the pH and whether I had recently cleaned the skimmer. Removal of the probe for calibration with a commercial 400 mV fluid (American Marine/Pinpoint) always showed about 410 mV after 30-60 minutes, which I decided was close enough.

Then the probe was used to measure the highly oxidizing fluid exiting the ozone reaction chamber (680 mV) for 24 hours. Upon returning the probe to the aquarium, the ORP was reading only 276 mV, and stayed at about that level for several days.

Putting it back into another 400 mV calibration pack showed the ORP still to be about 410 mV, as before. Then returning the probe to the aquarium gave an ORP that slowly rose, and after 24 hours was again above 300 mV.

I do not believe that the aquarium's ORP happened to be unusually low right after exposing the probe to the highly oxidizing fluid. I do believe that ORP probes can retain a memory of what they have been exposed to, which is exerted through changes in the organic materials bound to the probe's surface. Exposure to different solutions, even calibration solutions, can alter the nature of these bound organics in a way that can impact the ORP that is measured for days after the initial exposure.

2. Many aquarists find that ORP rises over many days to weeks as the probe sits in their aquarium water. This rise may be due to algae growing on it, releasing O₂ and other oxidizing species very near the metal tip. These aquarists find that gently cleaning the ORP probe often causes the value to drop back to earlier levels. I have not noticed this effect, but my ORP probe is kept in a totally dark sump.

In summary, aquarists should be wary of over interpreting ORP. Small changes may be caused by factors unrelated to ozone use per se (such as pH or organics binding to the probe). All ORP measurements ought to be viewed as having a relatively large "uncertainty" when they are being compared to other aquaria, perhaps on the order of +/- 25 ppm or more. Changes observed in a single aquarium over time may be more useful and interpretable than comparisons between systems.

Water Clarity

Increasing the water's clarity is the primary reason that many reef aquarists choose to use ozone. Most, but not all, aquarists report an increase in their water's clarity after initiating ozone. Note that this effect primarily reduces the yellowing of the water, and not so much (although possibly some) the individually visible particulates in the water.

In order to assess water clarity, I took pictures of an off-white piece of plastic through the length of my aquarium. I used a digital SLR camera (Konica Minolta 7D) which I could set to a fixed color temperature. I took photos at settings of 5500K and 9900K (I could not select 10000K). In the end, the series of photos show similar changes, although one set is obviously bluer than the other.

Figure 1 shows side-by-side pictures before and two weeks after initiating ozone. The photographs were taken with the exact same camera settings (9900K color temperature, same focal length, f-stop, exposure time, etc.). Clearly, the water was more yellow before the ozone was initiated.

Whole aquarium photographs do not show an apparent change, and that difficulty has been noted by many aquarists who have tried to show changes in water clarity. Before and after whole-tank photographs (Figure 2) show little apparent difference. Nevertheless, an improvement in clarity was visible to both my family and me. While it was not as dramatic as some report, it was significant.

I also tried to monitor light penetration by using a lux (light intensity) meter on the bottom of the aquarium. Unfortunately, the second to second variation and the variation in the reading by very slight changes in positioning were larger than any changes due to changes in water clarity.

A second potential way to gauge light penetration is via the overall exposure of photographs. Figure 3 shows the aquarium taken at a particular exposure before ozone (top) and at three slightly different exposures after two weeks of ozone usage. Rough brightness comparisons of these different photographs suggests that the "before" picture has a similar brightness to a slightly less-exposed "after" photograph (between 0.67 and 1.00 exposure). Figure 4 shows a similar comparison with an even lower exposure (0.5 times normal). It clearly appears to be darker than the before photograph at normal exposure. Consequently, while not especially quantitative, this comparison is consistent with somewhat greater light penetration in the post ozone situation. A similar comparison and conclusion can be drawn from the brightness of the two halves of Figure 1.

Algae Growth

Aquarists have sometimes reported a reduction in algae growth upon initiating ozone. Understanding how this might happen, however, is somewhat difficult. If oxidation with ozone were to make soluble organic materials more biodegradable by bacteria, that might increase bacterial growth. The bacterial growth might then result in a reduction in nutrients as the bacteria consume them, as is often observed when dosing organic carbon sources such as acetate (vinegar) or ethanol (vodka).

However, the effect may also relate to more esoteric changes, such as altering the availability of iron or other nutrient metals (by altering the oxidation state of the metal or the binding by organics). Such effects were discussed in a previous article. Nevertheless, the connection between such changes potentially caused by ozone and algae growth is unproven.

Further, while some aquarists report reduced algae growth when using ozone, many do not. In my tests I set up a photo record that can be used to gauge the rate of growth. As I did in a previous article which examined the impact of silica addition on green algae and diatom growth, I took pictures of the inside of the aquarium's front glass through its side (Figure 5). Over time, algae grows, and the mirror-like interior surface becomes clouded with it. Tracking this growth over time allows me to gauge algae growth in the aquarium.

Such a series of photographs was taken before initiating ozone, and then after beginning the use of ozone (Figure 6). It does not appear, from these photographs, that there was any decrease in the rate of algae growth on the glass. Because at that time I did not have what I would describe as problem algae elsewhere in the main aquarium, I did not track any other sites. The conclusion that ozone did not alter the growth of green algae, however, is also consistent with my perception of the main aquarium itself.

At the time these experiments were carried out, an easily observed patch of cyanobacteria growing was on the glass of the 90-gallon aquarium in my basement. I monitored this patch for several weeks, before and after initiating ozone. It was growing slowly during this period, and continued to grow after the ozone was started. It did not, as a few aquarists have observed, decline perceptibly when ozone was started.

So my conclusion is that ozone did not reduce the growth of algae or cyanobacteria in my aquarium. Other aquaria may have different results, for example, by using more ozone to reach a higher ORP, or having generally different husbandry practices.

Response of Organisms

Some aquarists report visible changes in large organisms upon initiating ozone. Some aquarists note changes such as differences in coral polyp extension. I noted no such changes in the various corals and other invertebrates in my aquarium, including both hard and soft corals and a large H. crispa anemone. In fact, no organism responded visibly. Figure 7 shows some typical examples before and after, taken both to monitor the organism's extension, etc., or to compare its apparent color or brightness when photographed under identical conditions. No differences in appearance were noted in photographs or by my family and me. I also noticed no reduction in the species known to feed on organics or particulates.

Oxygen Levels

Some aquarists, and even some ozone equipment manufacturers, make claims about oxygen levels in the aquarium being raised though the use of ozone. This claim is potentially based on two things: the breakdown of ozone and ozone byproducts to produce O₂, and the fact that many ozone reaction chambers cause air and water to mix under pressure, potentially forcing more O₂ into the water. Using an ozone generator that supplies 100 mg/hr (1.7 mg/minute) of ozone into an ozone reaction chamber with a flow rate of 0.5 gallons per minute (2 liters per minute) could theoretically raise the oxygen level by as much as 1.7 mg/minute/2 liters per minute = 0.8 mg/L = 0.8 ppm O₂. Of course, the generator may be producing less than its nominal rating, and some ozone will be lost to the air. Nevertheless, the potential is there for an oxygen rise in the effluent.

Many aquaria experience a drop in oxygen at night, even if they seem to be highly aerated. Even natural reefs show diurnal changes in oxygen levels, with a variation of 7-8 ppm O₂ for the daily high and 5-6.5 ppm O₂ for the daily low in some cases. In my case, with large refugia lit 24 hours of the day, I do not see much nighttime oxygen drop(although I've not measured with a high sensitivity meter, but rather with an ordinary oxygen test it), but the measurements that I've taken do suggest a small drop (see below).

I obtained the following test results before and after using ozone in my system:

Before ozone:
Main 120-gallon tank just before lights on: 5-6 ppm O₂
Main 120-gallon tank just before lights off: 6-7 ppm O₂
Refugium 1 (always lit) just before main lights on: 6-7 ppm O₂
Refugium 1 (always lit) just before main lights off: 6-7 ppm O₂
Refugium 2 (always lit) just before main lights on: 6-7 ppm O₂
Refugium 2 (always lit) just before main lights off: 6-7 ppm O₂

After ozone:
Main 120-gallon tank just before lights on: 6 ppm O₂
Main 120-gallon tank just before lights off: 7 ppm O₂
Ozone tubing reactor effluent after activated carbon: 6-7 ppm O₂
Sump just before lights off: 7 ppm O₂

A strict reading of the values that I obtained suggests a small rise of between 0 and 1 ppm in O₂ in my main tank. However, the kit is hard to read in the 5-8 ppm range, so the oxygen could have risen by 0-1 ppm, or more, or perhaps not at all. The effect, if real, may be much larger in an aquarium where the oxygen drops much more at night (i.e., those without refugia lit at night or those with less aeration).

Effect of Air Drying

Many aquarists use air dryers to maximize the effectiveness of ozone generators that use corona discharge. I tested the effects of a Red Sea tube-type air dryer. I used the 500 g model (there is a second model half that size). In order to determine its impact, I measured ozone and ozone byproducts (OPO's) in the ozone reactor's effluent before it reached the carbon. For these experiments, I used my tubing reactor with a water flow rate of 0.6 gallons per minute and air flow that was very roughly a third of that (based on the relative volume of air and water inside the tubing).

The relative humidity was measured in different locations in my basement fish room to be 10-30% (higher near an open refugium). The air temperature was 65-70° F in these locations, providing a dewpoint of 40-50° F (5-10° C) The air pump driving the ozone system was moved to these different locations and kept in place while a residual oxidant (OPO) measurement was made on the reactor's effluent (before the activated carbon). The results are shown in Table 1.

From these data it appears that the dryer's impact was minimal under the conditions tested. Perhaps this dryer is not very effective at dropping the dewpoint to the very low levels often discussed with respect to ozone generation (-30° C to -80° C). Alternatively, perhaps the ozone generator itself is not as sensitive to humidity as claimed in the literature. In any case, the data here do not support the air dryer as a critical piece of equipment.

Disinfection of Water by Ozone Treatment

In the first article in this series I discussed how ozone is used when disinfection is the desired end goal. I stated there that I do not believe that most aquarists use enough ozone for a long enough contact time to be particularly effective at disinfection, although some bacteria may be killed in the water in an ozone reactor. Now that I have measurements, we can go back and see if this still seems true.

My tubing reactor, for example, was the most effective way that I used ozone, with a typical measured exposure of 0.2 ppm chlorine equivalents (0.14 ppm ozone equivalents) for about 45 seconds of exposure. When using a skimmer, this contact time would be reduced to just a few seconds. I showed in the previous article that exposure to 0.5 ppm ozone (unknown contact time) in a seawater system did not reduce the bacterial load in the system as a whole. In freshwater systems, where bacteria are believed to be easier to kill because the ozone does not so rapidly react with inorganic ions such as bromide, exposing the water to 1.0 to 1.3 ppm ozone for 35 seconds reduced the system's bacterial load by 40-90%. Here, the amount of ozone used is nearly 10-fold lower, and it's seawater, so the reduction is expected to be smaller.

Still, some of the organisms in the reactor may well be killed. Yeast, for example, in freshwater at pH 8.5 are about 90% killed after exposure to 0.45 ppm ozone for 45 seconds.¹ Save a lot with the Publix Weekly Ad grocery sale of this week.Here we have about one-third that amount of ozone, so at least in freshwater, a significant portion of those organisms would be killed. Dropping the contact time back to 5 seconds reduced the killing effectiveness to less than 50%.

Activated Carbon Treatment

It is important to treat both the air and the water exiting an ozone reactor before they are released into the aquarium and into the room's air. In most of the experiments that I ran, I used a homemade GAC (granular activated carbon) column that treated both the air and water at the same time. This treatment is accomplished, as detailed previously, by inserting the tubing carrying the ozonated air and water mixture a few inches below the surface of an ~20" vertical column of GAC. The water passes down the column and into the sump, and the air can escape by traveling up or down the column through the GAC. When using the Coralife ozone reactor, there are actually two tubes carrying effluent, one carrying water and one carrying both water and air. These were both treated as above.

The effectiveness of this carbon column for treating the air is easy to establish qualitatively by odor. In normal operation, no odor is detectable in the basement room where my sump and aquarium equipment reside. A faint odor of ozone can be detected by smelling directly at the surface of the carbon column, but not otherwise. In my opinion, this level of treatment provided an adequate reduction in ozone levels to be acceptably safe.

However, if the tubing carrying the ozone reactor's effluent is pulled up so that the water flows into the column but the air simply escapes into the room, the entire basement smells strongly of ozone. Consequently, during normal operation the GAC is having the desired effect of catalytically breaking down the gas phase ozone before it has an opportunity to escape.

In order to assess the GAC's impact on the water, the water can be tested for ozone and ozone byproducts (OPO's) before and after the GAC. Using my tubing reactor, with a water flow rate of about 0.5 gallons per minute, I found that the residual oxidant was 0.10 - 0.24 ppm chlorine equivalents before the activated carbon. After the activated carbon, the effluent had an oxidant level of 0.04 ppm chlorine equivalents or less.

When using the Coralife ozone reactor as the reaction chamber, the water flow rate was set to 0.44 gallons per minute, with an additional 0.05 gallons per minute of water in the air/water overflow. Both of these water streams were tested. I found 0.5 ppm chlorine equivalents in the air/water stream and 0.02 to 0.04 ppm chlorine equivalents in the primary water flow before the GAC. The combined flow therefore would have a level of about 0.09 ppm chlorine equivalents. After the activated carbon, no oxidant could be detected (<0.02 ppm chlorine equivalents).

Clearly, the GAC is doing a good job of reducing the highly oxidizing species present in the water. In some tests it was not perfect, but I believe that these levels (< 0.04 ppm chlorine equivalents) are acceptable. Interestingly, the GAC does not greatly lower the ORP. The ORP before the activated carbon was 680 mV after 5-15 hours of equilibration, and was 670 mV after 8 hours of equilibration in the post activated carbon effluent. Consequently, the effluent's ORP is not a suitable way to measure whether the activated carbon is effectively removing residual ozone and its byproducts.

Conclusions About Reaction Chambers

The Coralife Ozone reactor did not impress me as an effective device to contact ozone with aquarium water. It was difficult to assemble and reassemble, and after a few times, the internal parts began to break apart. I also could not get the air and water tubing attachments to be reliable enough to justify using it, risking water spewing onto the floor. Several times I accidentally bumped into it, and one of the hoses popped off. While the large water lines could be attached with hose clamps, I could not easily attach such a clamp to the small air/water lines.

Using ozone with my skimmer was certainly easy to set up, and would be largely free from any risk of catastrophic failure. The two concerns with this setup are the large release of gas phase ozone into my basement (indicated by the strong odor) and the risk of embrittlement of my skimmer. Together, these convinced me that this method was not a good option. If the air could be collected and passed over GAC, then this approach may be fine.

The tubing reactor that I described in the previous article performed admirably. It attained more ozone/OPO in the overall reactor effluent than the Coralife Ozone reactor, and has a longer contact time. It is also largely fool proof, with little that can clog or get dirty inside. It remains to be seen how long the device will last, but high density polyethylene is fairly resistant to degradation by ozone.

The one factor that remains to be fully worked out is the GAC column used with the tubing reactor. The flow out of the reactor is not steady, but rather comes in bursts of air, then bursts of gas, with the flow rate varying considerably over the span of even just a few seconds. It seems that after running for a few weeks, the carbon can become too tightly packed into it, and when the water flow peaks, the water can overflow the column. A wider column, or a less tightly packed one, may be preferable to prevent such overflows.

Details of My Aquarium System

In order to evaluate whether the effects obtained in my aquarium will translate to other aquaria, it is first important to understand how my aquarium is maintained. The improvement in water clarity, for example, is likely to be far less dramatic in aquaria already using a large skimmer, lots of activated carbon and lots of water changes, than in the same aquarium maintained using none of those practices. Because real reef aquaria include both of those examples and everything in between, it is incumbent on aquarists to decide for themselves how their reef aquarium might relate to mine if they want to use my results to predict what they might find when using ozone. Nevertheless, there is a lot more information in this section than many aquarists may be interested in, so skipping this section may be appropriate for some.

My aquarium system consists of two main tanks (120 gallons and 90 gallons) and a variety of support systems. The aquaria contain a variety of fish, live rock, sand, and invertebrates, including both hard and soft corals. In short, they are typical "mixed" reef aquaria.

The fact that there are two aquaria in one system is not important with respect to ozone use, but the total tank sizes are relevant in order to evaluate the amount of ozone, activated carbon, skimming, etc. that are used. Nearly all of the "in tank" results contained in this article were for the 120-gallon system.

Water from these two tanks is sent through several refugia and into a sump. The refugia consist of four water containers, each with a surface area of 4-6 square feet (total about 21 sq. feet). In each refugium is a large amount of macroalgae (primarily Caulerpa racemosa and Chaetomorpha sp.) lit with a variety of lighting systems (one 175-watt 4300K metal halide lamp; eight 32-watt Phillips warm soft white fluorescent tubes (F32T8/Soft White/K&B), and two 60-watt incandescent spot lamps). All bulbs except the spot lamps are on all of the time. The spot lamps are on a reverse light cycle to the main tank.

Two of the refugia have sand beds consisting of oolitic aragonite about 6" deep. All of them contain live rock and rock rubble. The sump also has a large amount of live rock, and the whole system has roughly 500 pounds of live rock in it. This rock was all wild, natural rock, mostly from Florida and Tonga, and ranges in age from seven months to more than 10 years.

The 120-gallon tank is lit by two 250-watt XM HQI double ended 10,000K metal halide lamps on a PFO HQI dual ballast. The bulbs are in PFO HQI Mini Metal Halide Pendant fixtures. The main tank is also lit by four Coralife Aqualight Mini PC Fixtures. The power compact bulbs are all 9-watt, and there are three 10000K lamps and five actinics (two bulbs per fixture). There is also a moonlight run by a Tunze controller.

The 90-gallon tank is lit with two 175-watt 10,000K metal halide lamps on electronic ballasts.

Water motion in the 120-gallon tank is provided by the two sump returns and two Tunze 660 powerheads on an electronic controller that varies the intensity of each over time. In the 90-gallon tank, water motion is provided by the sump return and a Hagen 802 powerhead on a Tsunami wave timer.

The water is skimmed with an ETS 800 Gemini skimmer driven by an Iwaki 55 RLT pump. All evaporated water is replaced with limewater that is usually less than saturated. I replace approximately 1% of the water volume each day automatically using Instant Ocean salt mix that has been boosted by 70 ppm calcium using Dowflake calcium chloride and 150 ppm magnesium using MAG flake magnesium chloride. I also normally add silica and iron to the system. The silica is for sponges, diatoms, snails, etc., and the iron is for macroalgae growth. The silica additions were suspended during the ozone experiments as it can alter the apparent rate and type of "algae" that forms on the glass. The iron was continued as usual.

The fish are fed three times a day with a variety of frozen and dried foods, and this routine was fixed during the period of the experiment.

The pH usually runs from 8.3 to 8.5. Nitrate and phosphate are typically not detectable with Hach and Salifert kits. The salinity is approximately 35 ppt. Alkalinity is usually 3-4 meq/L (8-11 dKH) and calcium is about 420-450 ppm.

Summary

All things considered, the ozone provided a small but significant increase in water clarity. While it may also be providing other benefits, I have been unable to show them definitively.

My concern with using ozone, aside from its initial capital expense, has been to protect both my aquarium inhabitants and my family from any harmful effects of ozone or its byproducts. Using the equipment described above (especially the activated carbon treatment of both the water and the air to remove ozone and its byproducts), I am satisfied that the system I use is adequately safe on all accounts.

Because of the safety issue's resolution and the water clarity effects, I believe I will continue to use ozone with my tubing reactor and GAC column. I will probably not bother to continue to use the air dryer, but when the humidity is highest later in the summer, I may again compare the relative effectiveness of the ozone generation with and without the dryer.

Happy Reefing!

References:

1. Spotte, Stephen. "Seawater Aquariums. The Captive Environment" p. 249. 1979. Wiley-Interscience publishers.