One of the most common concerns of reef aquarists is pH. Some aquaria have pH that is too high, and some too low. Some have the pH just right, but the aquarist thinks it isn't. In a previous article on solving pH problems, and in a second one specifically directed to diagnosing and solving low pH problems, I pointed out that the first step is to ensure that the pH is being measured correctly. The two primary tools for measuring pH in aquaria are pH kits and pH meters. In an earlier article, I described in detail how pH meters work and how to use them. This article expands on that article by reporting the results of tests on a variety of commercial pH meter calibration buffers to determine their suitability for reef aquarists.

Unfortunately, as the results below make clear, not all commercial calibration buffers used by reef aquarists are suitable. The inaccuracies in at least one of the brands is so extreme as to make it worse than useless, potentially driving aquarists to 'solve' problems that do not exist, and possibly moving their aquaria's pH to undesirable levels.

Contents:

• What is pH?
• Why Monitor pH?
• Acceptable pH Range
• Testing pH Calibration Buffers
• Interpreting pH Test Results
• Expiration Dates on Calibration Buffers
• "Calibration" Using Borax
• How to Calibrate and Use a pH Meter
• Summary

What Is pH?

The chemical parameter referred to as "pH" is a measure of aquarium water's acidity. The concept of pH in a seawater application has a variety of different definitions. In the system most aquarists use (the NBS system, with NBS standing for the old National Bureau of Standards), pH is defined in equation 1:

1.  pH = -log a_H

where a_H is the "activity" of hydrogen ions in the solution. Activity is how chemists measure "free" concentrations. So pH is simply a measure of the hydrogen ions (H⁺; protons) in solution. In order to understand most pH problems in marine aquaria, however, the difference between activity and concentration can be ignored, and pH can simply be thought of as relating directly to the concentration of H⁺:

2.  pH = -g_Hlog [H⁺]

where g_H is simply a constant (the activity coefficient; g_H = 1 in pure fresh water and ~0.72 in seawater) that we can also ignore for this purpose.

In a sense, all that most aquarists need to know is that pH is a measure of the hydrogen ions in solution, and that the scale is logarithmic. That is, at pH 6 there is 10 times as much H⁺ as at pH 7, and that at pH 6 there is 100 times as much H⁺ as at pH 8. Consequently, a small change in pH means a big change in the concentration of H⁺ in the water.

Why Monitor pH?

There are several reasons to monitor pH in marine aquaria. One reason is that aquatic organisms thrive in only a particular pH range. This range certainly varies from organism to organism, and it is therefore not easy to justify a claim that any particular range is "optimal" for an aquarium with many species. Even natural seawater (pH = 8.0 to 8.3) isn't going to be optimal for every creature living in it, but it was recognized more than eighty years ago that moving away from the pH of natural seawater (down to 7.3, for example) is stressful to fish.^1,2 There is now additional information available about optimal pH ranges for many organisms, but the data are woefully inadequate to allow aquarists to optimize pH for most organisms in which they are interested.^2-6

Additionally, the effect of pH on organisms can be direct, or indirect. For example, the toxicity of metals such as copper and nickel is known to depend on pH for some of the organisms present in our aquaria.⁷ Consequently, the ranges of pH that are acceptable in one aquarium may be different in other aquaria, even for the same organisms.

Nevertheless, some fundamental processes taking place in many marine organisms are substantially impacted by changes in pH. One of these is calcification, and it is known that calcification in corals is dependent on pH dropping as the pH falls.^8,9 Using this type of information, along with the integrated experience of many hobbyists, we can develop some guidelines about what is an acceptable pH range for reef aquaria and what values are considered to be "pushing the limits."

Acceptable pH Range

The acceptable pH range for reef aquaria is an opinion rather than a clearly delineated fact, and will certainly vary based on who is providing the opinion. This range may also be quite different from the "optimal" range. Justifying what is optimal, however, is much more problematic than that which is simply acceptable, and we will focus on the latter. As a goal, I'd suggest that the pH of natural seawater, about 8.2, is appropriate, but aquaria can clearly operate in a wider range of pH values. In my opinion, the pH range from 7.8 to 8.5 is an acceptable range for reef aquaria, with several caveats. These are:

1. That the alkalinity is at least 2.5 meq/L, and preferably higher at the lower end of this pH range. In part, this statement is based on the fact that many reef aquaria operate acceptably in the pH 7.8 to 8.0 range, but that most of the best examples of these types of aquaria incorporate calcium carbonate/carbon dioxide reactors that, while tending to lower the pH, keep the carbonate alkalinity fairly high (at or above 3 meq/L.). In this case, any problems associated with calcification at these lower pH values may be offset by the higher alkalinity.

2. That the calcium level is at least 400 ppm. Calcification becomes more difficult as pH drops, and it also becomes more difficult as the calcium level is lowered. It would not be desirable to push the extremes of pH, alkalinity, and calcium all at the same time. So if the pH is on the low side and cannot be easily changed (such as in an aquarium with a CaCO3/CO2 reactor), at least make sure that the calcium level is acceptable (~400-450 ppm).

3. Likewise, one of the problems at higher pH (above 8.2, but becoming progressively more problematic with each incremental rise) is the abiotic precipitation of calcium carbonate (resulting in a drop in calcium and alkalinity, and the clogging of heaters and pump impellers). When the pH is 8.4 or higher (as often happens in an aquarium using limewater), make sure that both the calcium and alkalinity levels are suitably maintained (that is, neither too low, inhibiting biological calcification, nor too high, causing excessive abiotic precipitation on equipment).

Testing pH Calibration Buffers

In order to accurately use a pH meter to measure pH, it MUST first be calibrated with solutions of known pH. Some tips on how best to accomplish this task are given later in this article. One presumption in calibrating a pH meter is that the pH calibration buffers being used are accurate. Unfortunately, commercial buffers do not always meet this need for accuracy.

In order to test some of the commercial buffers most often used by aquarists, I organized a test of many of the brands most commonly used by aquarists. I purchased some of these from a local fish store (LFS), and others from three of the largest online vendors of aquarium supplies. In most cases, the buffers were purchased from two or three different suppliers to ensure that the results reflect what is generally available, and to avoid specific vendor problems (such as old products). I also tested some laboratory brands that I obtained from a laboratory supply company (LSC) as well as some unopened, but old (expired), pH calibration buffers. Issues with expiration dates are discussed later in the article.

I have chosen to show the names of the brands that appear to be sufficiently accurate for typical reef aquarium purposes. I have chosen not to name two brands that I deemed not suitably accurate for use as a standard. One of these was grossly inaccurate, and could lead to serious problems if its buffers were used to determine and subsequently "correct" the pH in a reef aquarium (or in a CaCO3/CO2 reactor).

To perform the tests, I used a Chemcadet pH meter/controller from Cole Parmer (Figure 1). The pH electrode that I used was a "new" (meaning unused, but manufactured a few years ago) Orion 9256 BN which has an epoxy body and a Ag/AgCl internal reference electrode (Figure 2). It was filled with 4 M KCl saturated with AgCl and soaked in diluted reef aquarium water for three weeks before use. All tests were performed at 22-23ºC. The temperature setting of the pH meter was set and left at 22ºC throughout the testing.

One issue that arises when testing pH buffers is how to ensure that the meter is correctly calibrated in the first place. I chose to use individual packets of pH 7 and 10 calibration buffers from a highly trusted laboratory instrument company, Thermo (previously known as Orion). I used these buffers to calibrate the meter, as shown in the top two lines of Table 1. The temperature was maintained at 22-23ºC which, according to the table on the buffer packet, ought to yield pH values of about 7.00 and 10.04, and that is the value to which the meter was calibrated.

In retrospect, this calibration method appears to have been an acceptable choice, as most of the other brands matched the results obtained with this calibration quite well (shown in green in the tables). I consider a difference of about 0.1 pH unit to be unimportant to a reef aquarist, while a difference of 0.1-0.2 is becoming significant. A difference of more than 0.25 pH units is too much, in my opinion. The worst samples were far outside the envelope of possible random error (shown in red in the tables).

The testing of all of these samples took a few hours. During that time, it is certainly possible that the pH meter drifted. In fact, I believe that it did, as shown below. For that reason, the data in Tables 1 and 2 are shown in the chronological order in which the tests were taken. About midway through the test period, the pH values of the original pH 7 and 10 calibration fluids were remeasured (reading pH 6.92 and pH 9.92). While I could have recalibrated the meter, I chose not to since the calibration fluids may have changed upon exposure to the air (not likely, but possible). Instead, the measurements taken near the end of Table 2 are likely a bit lower than the actual values, and any interpretations should take that likelihood into account. Overall, the errors that would concern a typical aquarist are larger than this level of drift, and I considered it to be unimportant to the final analysis.

Interpreting pH Test Results

Based on the results in Tables 1 and 2, I conclude that several brands are suitably accurate for reef aquarists (with a typical variation of less than 0.1 pH unit), and these are the brands (in no particular order) that I recommend:

Recommended Calibration Buffers: Thermo (previously known as Orion) Hanna Milwaukee Pinpoint (made by American Marine)

I have decided that two of the brands I tested are not suitably accurate, and decided not to reveal their names to avoid the heated debate that might occur if I did. To avoid them, simply buy one of those recommended above. I expect that the Corning brand (a high quality laboratory pH meter maker) is acceptable, but I did not test new samples in the pH range of interest. The purpose of testing the Corning samples was to see how the pH 4 samples changed with advanced age (and the fact that I had such old samples of only that type, but no other).

How serious are the worst? Very. Even if we ignore the incredible pH 4 buffer with a pH of 6.1, we still have the following possibilities using the actual measurements taken above:

Suppose that an aquarist calibrated with a pH 7 buffer that was really pH 7.5 and a pH 10 buffer that was really pH 10.0. In that case, the natural seawater pH value of 8.2 would read as pH 7.8.

Suppose that an aquarist calibrated with a pH 7 buffer that was really pH 7.0 and a pH 10 buffer that was really pH 9.01. In that case, the natural seawater pH value of 8.2 would read as pH 8.8.

Suppose that an aquarist calibrated with a pH 4 buffer that was really pH 4.58 and a pH 7 buffer that was really pH 7.0. In that case, the natural seawater pH value of 8.2 would read as pH 8.5.

Suppose that an aquarist calibrated with a pH 4 buffer that was really pH 4.0 and a pH 7 buffer that was really pH 7.5. In that case, the natural seawater pH value of 8.2 would read as pH 7.6.

It is clear that something is seriously wrong if the values for natural seawater with a real pH of 8.2 can range from pH 7.6 to pH 8.8 when using instruments calibrated with these commercial buffers.

Expiration Dates on Calibration Buffers

In general, expiration dates on calibration fluids are desirable. The pH 10 calibration fluids are known, for example, to absorb carbon dioxide from the air and to drop in pH over time. A sealed foil packet may be fairly slow to absorb carbon dioxide and drop in pH, but without a date, it is impossible to know how old it is. It might be five years old, or more. The expired Milwaukee bottles in Table 1 (expired in April 2004, bought in December 2004, tested in January 2005) were bought at a local pet shop (at a discount due to the expiration). The same brand in a foil packet bought in December 2004 from an online retailer was dated to expire in February 2006. I do not know if the bottle initially had a longer or shorter shelf life than the same fluid in a bottle, but if similar to the packet, it would have been sitting around for almost two years longer than the foil packet.

Also note that the Hanna pH 10 bottle that I bought has an expiration date more than 4 ½ years into the future (June 2009; bought December 2004). That seems excessive to me, but at least a date is present and they hopefully have reason to believe that it is stable that long.

All of the samples that had dates and that were not expired were acceptable. Even expired versions of some samples were acceptable, including the Corning samples that were more than eight years old. That result may reflect the fact that pH 4 buffers do not suffer as greatly from carbon dioxide absorption as do higher pH solutions. It may also reflect the quality of the packaging. I also had 13 individual plastic packets of VWR brand pH 7 buffer, none of which still contained any liquid. They expired in 1996, and the liquid in them all apparently evaporated over the years.

In summary, however, the brands that failed had no expiration dates, while all of those that did have such (unexpired) dates were acceptable.

"Calibration" Using Borax

As shown above, commercial calibration solutions can be inaccurate. Aquarists who are concerned about accuracy can verify proper operation by testing their meter in other standard solutions. One such solution is borate, at about pH 9.2. Craig Bingman described that useful test in a previous article. In that instance, a pH 9.2 solution is made by dissolving a measured amount of borax (from a grocery store) in pure water.

Many meters cannot be truly calibrated with this solution, since they may require that the buffers be pH 4, 7, or 10. If a meter allows input of other pH values, then the pH of 9.2 that results from borax in water can be used for calibration. In all cases, though, it can be used to verify proper operation after calibration. Even this method is not foolproof, however. The water used to make the borax/water solution may not be adequately pure, or too much CO2 may enter the solution from the air. Nevertheless, it is a good check for aquarists concerned about the accuracy of the pH measurement.

How to Calibrate and Use a pH Meter

The most important aspect of using a pH meter is correctly calibrating it. Each meter will have a slightly different method of calibration. A number of general rules are very useful, however:

1.  Any analytical method, including pH measurement, is best calibrated with quality standards that span the range expected to be measured. Most aquarists calibrate pH meters using two solutions of known pH. A meter that allows only one calibration point is a very poor choice. Using more is fine if the meter allows more than two. When using two solutions to calibrate a pH electrode for use in a marine aquarium, one calibration point should optimally be below 8.0 (typically 7) and one should be above 8.5 (typically 10, but 9 is also sometimes used). When measuring pH in something other than aquarium water, there may be special tricks to use which are detailed below.

Using pH 4 and 7 is often done, but can be a less optimal choice because the range expected to be measured for reef aquarium water (about pH 7.8 - 8.6) is outside of this calibration range. In some cases the error is small enough that this is acceptable; while in others, it may be a problem.

The tables below show the maximum errors that are attained from various errors in the standard buffer solutions themselves (with problems with the standards being the only source of error considered; in reality, there can be additional errors in real measurements). These tables were obtained by simply looking at how much the calibration solutions might vary (first column), then seeing by how much the actual measured value can be off if both standards vary to the stated maximum error and in directions that result in the maximum measurement error (which turns out to be varying in opposite directions when using pH 4 and 7, and varying in the same direction when using pH 7 and 10 standards).

It is clear that with similar errors in the standard solutions, the errors in the measurements at pH 8-10 are smaller when calibrating at pH 7 and 10 than at pH 4 and 7. Whether these differences are important depends on the application and expectations of the aquarist.

Additionally, when measuring pH in a fluid of a lower pH (such as inside a CaCO3/CO2 reactor), calibrating at pH 4 and 7 is more sensible than calibrating at pH 7 and 10.

2.  Make sure the calibration standards are either new or at least adequate for the purpose. To be sure, use one of the brands recommended in this article.

I have several bottles of pH fluid that I have been using for years. Occasionally, I use a fresh bottle or packet to calibrate my meter. At that time I check the pH values of all of these older bottles, and note the pH on the bottle. I can then use that bottle for future pH calibrations BECAUSE my meter allows me to calibrate with standards at any pH (such as pH 7.03 and 8.85). If the meter does not allow the input of the pH values that precisely, then it is not possible to use this trick.

3.  Rinse the pH electrode in pure fresh water before putting it into any calibration standard, and between each standard. Also, be careful to not transfer anything except a trace of purified water into the calibrations standards. Even tap water, when transferred into a calibration standard, can impact the pH.

4.  It takes some time for a pH meter to get a correct reading. So let the meter equilibrate to each standard long enough that the value stabilizes (say, holding within +/- 0.02 pH unit for 30 seconds or longer). Some meters beep or otherwise indicate when they are suitably equilibrated.

5.  Stirring the solution can help the pH probe equilibrate to the solution, but it also encourages CO2 to enter the fluid. This CO2 can lower the pH of high pH standards, such as pH 8 and greater. I stir mine for about 30 seconds (often with the pH probe itself, though I've also broken them this way) and then let it sit to get a reading.

6.  The temperature of the standards is important for two reasons. One is that temperature changes actually change the standards' pH. The other is that the pH electrodes change their response as a function of temperature (described above). The change in standard solution pH as a function of temperature cannot be automatically adjusted for by inputting temperature into the meter, or via its ATC. It is an attribute of the exact chemistry of the buffer used. Some have pH that rises as temperature rises, and some fall as temperature rises. Others rise with temperature in some temperature ranges and fall with temperature in other temperature ranges. Aquarists should be aware of the exact pH at the temperature being used for calibration. Buffers will often have such pH values as a function of temperature printed on the bottle. For example, a standard phosphate buffer has a pH of 7.000 at 25ºC, but 7.04 at 15ºC (a small difference). At the same time, a carbonate buffer with a pH of 10.01 at 25ºC has a pH of 10.12 at 15ºC (a larger difference).

7.  After calibrating the meter, go back and make sure that it reads the calibrating solutions correctly (to within whatever error is acceptable) to be sure that it was done correctly.

8.  For certain kinds of pH measurements, direct comparison to a known standard may be more useful than using the absolute readings that the pH meter shows. For example, to assess the strength of limewater via pH, make a standard of known saturated limewater (from, for example, a teaspoon of calcium hydroxide in a cup of pure fresh water). That solution has a pH of about 12.45 at 25ºC, but regardless of what is measured, use the number as the standard and see how far off from actual limewater it is (if it is 0.1 pH unit lower, then the limewater is about 79% saturated; 0.2 pH units lower and it is 63% saturated; 0.3 pH units lower and it is about 50% saturated; 0.4 pH units lower and it is 40% saturated, etc.). In this case, exact temperature equivalence between the samples is important. A difference of only 3ºC means a pH difference of 0.1 pH unit for saturated limewater.

Summary

Measuring pH in some fashion is important for most reef aquaria. Using a properly calibrated pH meter is one of the easiest ways to accomplish this goal. Properly calibrating a pH meter, however, requires reasonably accurate pH calibration fluids. Many of those available to aquarists appear to have suitable accuracy, but some do not. In order to avoid problems with inaccurate brands, my suggestion is to use one of the brands found accurate in this article, or which are otherwise believed to be accurate. Those recommended in this article are:

Recommended Calibration Buffers: Thermo (previously known as Orion) Hanna Milwaukee Pinpoint (made by American Marine)

Special thanks to my laboratory assistant, Savannah Holmes-Farley (7 years old) who assisted by taking all the notes from these experiments.

Happy Reefing!

References:

1. Hydrogen-ion concentration of sea water in its biological relations. Atkins, W. R. G. J. Marine Biol. Assoc. (1922), 12 717-71.

2. Water quality requirements for first-feeding in marine fish larvae. II. pH, oxygen, and carbon dioxide. Brownell, Charles L. Dep. Zool., Univ. Cape Town, Rondebosch, S. Afr. J. Exp. Mar. Biol. Ecol. (1980), 44(2-3), 285-8.

3. Chondrus crispus (Gigartinaceae, Rhodophyta) tank cultivation: optimizing carbon input by a fixed pH and use of a salt water well. Braud, Jean-Paul; Amat, Mireille A. Sanofi Bio-Industries, Polder du Dain, Bouin, Fr. Hydrobiologia (1996), 326/327 335-340.

4. Physiological ecology of Gelidiella acerosa. Rao, P. Sreenivasa; Mehta, V. B. Dep. Biosci., Saurashtra Univ., Rajkot, India. J. Phycol. (1973), 9(3), 333-5.

5. Studies on marine biological filters. Model filters. Wickins, J. F. Fish. Exp. Stn., Minist. Agric. Fish. Food, Conwy/Gwynedd, UK. Water Res. (1983), 17(12), 1769-80.

6. Physiological characteristics of Mycosphaerella ascophylli, a fungal endophyte of the marine brown alga Ascophyllum nodosum. Fries, Nils. Inst. Physiol. Bot., Univ. Uppsala, Uppsala, Swed. Physiol. Plant. (1979), 45(1), 117-21.

7. pH dependent toxicity of five metals to three marine organisms. Ho, Kay T.; Kuhn, Anne; Pelletier, Marguerite C.; Hendricks, Tracey L.; Helmstetter, Andrea. National Health and Ecological Effects Research Laboratory, U.S. Environmental Protection Agency, Narragansett, RI, USA. Environmental Toxicology (1999), 14(2), 235-240.

8. Effects of lowered pH and elevated nitrate on coral calcification. Marubini, F.; Atkinson, M. J. Biosphere 2 Center, Columbia Univ., Oracle, AZ, USA. Mar. Ecol.: Prog. Ser. (1999), 188 117-121.

9. Effect of calcium carbonate saturation state on the calcification rate of an experimental coral reef. Langdon, Chris; Takahashi, Taro; Sweeney, Colm; Chipman, Dave; Goddard, John; Marubini, Francesca; Aceves, Heather; Barnett, Heidi; Atkinson, Marlin J. Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY, USA. Global Biogeochem. Cycles (2000), 14(2), 639-654.