Introduction
While pH is quite an important process variable and analytical parameter, there is much misunderstanding about what it is and much confusion when measurements go wrong. Without getting too deeply into theory, let's try to understand more about pH and its measurement.
We have a qualitative understanding of acids and bases from household experiences. You may have cleaned a patio or garage floor with muriatic acid (hydrochloric acid) and seen the bubbling reaction with the concrete. Lemons and limes are sour and oranges and grapefruit can be. We know pickles can be sour and we've seen the reaction between vinegar and baking soda. We've seen other leavening agents make cakes and biscuits rise up in the oven. We know there is something much stronger and reactive in drain cleaners that we pour or spoon into clogged drains.
There is also a quantitative aspect to acids and bases that allows us to say that muriatic acid is stronger than lemon juice which is stronger than vinegar. On the basic side, drain cleaner is clearly stronger than baking soda. These relative strengths come not just from our experience but also from measurements of pH of these various substances. The determination of pH allows us to compare various acids and bases in cases where we can't necessarily taste (generally unwise!) or see differences in reactivity.
First, let's establish that pH is not entirely a purely theoretical concept or treatment that can be entirely derived from theory. There are many conventions and compromises to make pH measurement, first of all, a practical matter. Then theory comes along behind to justify and explain what is being done rigorously in accord with theory and what is being overlooked as a concession to everyday practice. We'll try to distinguish between those points based on physical chemistry and those based on convention or compromise.
Secondly, let's distinguish between two ideas that are easily confused when talking about strong acids and strong bases. One idea is the amount of the substance or the concentration in the system. It is possible to have a strong acid in such low concentration that the acidity of the solution is almost inconsequential for many purposes. It is also possible to have high concentrations of a weak acid, but the solution will not be appreciably acid for many purposes. The second idea is that of acid strength. When placed in water, a weak acid does not dissociate, or split, into the charged species, called ions, to the same extent as a stronger acid. The hydrogen ion (or proton, hydronium ion) is the species responsible for acidity and the more there are in solution, the more acidic it is. So, the strength of the acid or base is quite a separate issue from the amount or concentration of that acid or base in the sample. It is a convention to reserve the terms strong acid or strong base for substances that dissociate completely into their component ions, and in an acid one component is the positively charged hydrogen ion (or proton, hydronium ion) and negatively charged anions; a strong base is one that dissociates completely into negative hydroxide ions and the accompanying cations. The term strong is used to define a molecular property, not as an indication of the amount of that substance.
Thirdly, let's understand that pH and the particular scale we use was developed to describe the behavior of aqueous solutions of acids and bases. That doesn't mean that pH hasn't been extended to other solvents or mixtures of solvents. But the details of our conventions go back to properties of water and aqueous solutions of salts, acids and bases. Trying to measure pH in non-aqueous solutions or even partially aqueous solutions can introduce problems and errors unless special precautions are taken. Avoiding these errors is part of our goal in this explanation of pH and its measurement.
Definition of pH
We talk about pH being a measure of acidity or basicity, but the definition only refers to the concentration of hydrogen ion, the species more associated with the concept of acidity. When this concentration is relatively high, the solution is acidic. When the concentration is low, the solution is basic. High or low compared to what? you might ask. Compared to water itself. There are several devices that can measure the concentration of hydrogen ions and when measuring pure water, they say 10-7 moles/liter is the concentration of hydrogen ions. Since pure water is about 55.5 moles/liter, this is the same thing as saying that one water molecule in about 78 million spontaneously splits into H+ and HO- ions. Thus, neutral, pure water has a measurable hydrogen ion concentration of 10-7 moles/liter. If some acetic acid is placed in the water, the concentration goes up to say 10-4 moles/liter. The hydrogen ion concentration has increased by a factor of 1000 due to this acidic substance. There is nothing subtle or difficult to detect about this difference and this is a fairly weak acid. Similarly, the base disodium phosphate can be added to the pure water and the hydrogen ion concentration might be easily measured at 10-8 moles/liter, a 10-fold reduction. The measurable range (by typical pH meter) of hydrogen ion concentrations spans from about 10 moles/liter to 10-14 moles/liter. This is a range of 15 orders of magnitude. Not many other common measurements cover such a wide range of concentration. Most techniques in analytical chemistry require adjustment of the sample (usually by dilution) to be in the working range of the instrument, but pH meters measuring aqueous solutions can handle the entire range directly. Rather than writing all these exponents for every measurement, the number one convention of pH is to express the hydrogen ion concentration as just the exponent or more precisely the negative of the exponent. The mathematical operation that does this is the negative of the common logarithm (base 10 logarithm, log10). Thus, 10-7 moles/liter, the neutral point with no added acid or base, becomes 7.00; an acidic solution of 0.25 moles/liter becomes pH 0.602 and a sample of laundry bleach of 4.0 x 10-11 becomes pH 10.40. This is the reason for the confusing convention of the pH number going down as the acidity and hydrogen ion concentration increase.
This definition of neutral pH being 7.0 only applies to 25°C. The same pure water with no (outwardly evident) change in composition when at 0°C will be measured as 3.3 x 10-8 moles/liter for a pH of 7.47 and at 60°C will be measured as 3.1 x 10-7 moles/liter or pH 6.51. This is due to a change in the dissociation constant of water with temperature. This behavior illustrates a problem with our current definition of pH, in that temperature and several other things all affect the outcome of our measurement. We need to recognize that we really aren't measuring concentration of hydrogen ions with the various pH electrode technologies available, but rather the concentration altered by a combination of known and unknown effects. This effective concentration that is actually measured is known as the activity of hydrogen ion in our sample, buffer or calibration solution.
Let's look at some of the parameters known to affect activity. (1) Temperature is important because chemical equilibria and reaction rates are affected directly by temperature. Even simple solutions have detectably different compositions at different temperatures. Measurements and calibrations must be at either conventional or specified temperatures so that another lab can reproduce the procedure. Next, (2) the charges on the ions in the solutions matter, as more highly charged ions (+2, +3, +4) attract and repel other charged species more energetically. (3) Ionic strength is a measure of the cumulative concentrations of all the ions from all the solutes in the solution. For example, NaCl at 1 mole/liter completely dissociates into 1.0 mole of Na+ and 1.0 mole of Cl- for an ionic strength of 1 moles/liter. Ionic strength depends on the square of the ionic charges, so ions with valencies of 2, 3 or 4 increase the ionic strength greatly. A different example of 1 M calcium chloride still has 1 mole of Cl- but the calcium cation has a charge of +2 and counts more in the formula, so the ionic strength is 3 mole/liter. (4) The dielectric constant of the medium (usually water, which at 78 is higher than almost all other solvents) indicates how effectively the charges are transmitted or shielded from each other. (5) The size of ions is a factor since the closer they can approach each other the greater the attractive or repulsive forces. (6) Another factor affecting activity is the density of the solvent. Several of these factors will be discussed later as we review specific problems and solutions. This area of physical chemistry is generally known as theory of ionic solutions and is described by Debye-Hückel theory which was introduced in the early 20th century and has been refined since.
The various contributors to activity are usually all lumped together in one factor called the activity coefficient. This is a multiplier, usually somewhat less than 1.0, that modifies the concentration by a factor of 1 - 30% or more. The main factor that controls the extent of deviation from ideality is concentration. The ideal behavior is described as infinite dilution in many treatments, indicating an imaginary system in which the solute molecules don't interact with each other and concentration and activity are equal. At 10-4 moles/liter, most solutes don't show much difference between concentration and activity, but at 0.1 moles/liter, most solutes do. At 1 mole/liter any ionic solute is clearly far from ideal and many factors are operating to reduce the effective concentration.
Measurement of pH
Determining pH of a sample requires three basic components: a pH-sensing electrode, a reference electrode that are both exposed to the sample solution and electronic circuitry that measures the potential between these two electrodes and translates it into pH units. Each of these functions has its own idiosyncrasies and precautions that need understanding and attention by the operator of the equipment. We'll discuss each of these in turn and how these details lead to practical precautions that should be observed.
The Hydrogen Electrode
While glass electrodes are the standard pH devices now, the hydrogen electrode was developed first and is still the standard electrode to which potentials in other electrode types are referenced. It is not complicated to construct as a laboratory reference, but it is inconvenient to use, chemically incompatible with many samples and not without hazards. It consists of a catalytically active metal such as platinum or palladium that will convert H2 gas molecules to hydrogen ions and electrons that also serves as an electrode for electrical connections. Hydrogen gas is bubbled through the aqueous solution in which the electrode is immersed. A second electrode that is not catalytically active, such as silver/silver chloride or calomel, is in contact with the same sample solution to provide a reference (zero) potential and complete the circuit. The potential generated is proportional to the log of the hydrogen ion concentration of the aqueous solution in the cell, as well as the partial pressure of hydrogen applied to the cell. A number of other minor factors are involved and adjusted when precise work is performed. Clearly, this electrode is not compatible with solutes that are easily oxidized or reduced and a stream of flammable hydrogen gas is generated. Any number of substances (sulfur compounds, etc) can poison the catalytic surface. Equilibria in dilute and weakly buffered solutions are disturbed by the normal operation of this electrode. It is therefore used only for well-designed fundamental studies of pH and as a primary reference for glass electrodes in laboratories that have such a need. The standard hydrogen electrode (SHE) has its own conventions in which actual parameters from the measurement are corrected to a defined standard state. Its ease of construction and facile generation of reference potentials makes it a standard to which other devices are referenced.
The Glass Electrode
The glass electrode was developed at a time when the concept of pH was being developed and the limitations of the hydrogen electrode were quite clear. An observation was made that a thin glass bulb of the proper composition developed a potential depending on the acidity of the solution. This started decades of study into the details of the glass and internal components needed to reproducibly measure pH. The field of ion-selective electrodes also developed concurrently with glass pH electrodes so that understanding one helps in understanding the other. Today, most pH electrodes are combination electrodes that contain the pH-sensing element, the reference electrode and often the temperature sensor in one body. It is important to understand the separate functions of each and how each can go wrong or induce errors.
The simplest pH-sensing glass electrode is a bulb of the proper type of glass filled with a conducting aqueous solution of KCl or other electrolyte into which is immersed a wire or conductor. The following discussion shall refer to the diagram illustrating these parts. The glass composition is important since there are many glasses that are not pH-sensitive. An early standard pH glass is known as Corning 015 and is useful for illustrating features and improvements in glass electrodes. Corning 015 consists of 72.2 mole% SiO2, which is expected in glass, but also 21.4 mole % Na2O and 6.4 mole % CaO which make the glass useful in pH work. The features of the inside surface, inner solution and internal conductor do not change or enter into the discussion nearly as much as the outer surface of the glass bulb. Placing the bulb into a solution of aqueous acid induces a potential (voltage) across the two surfaces of the glass. In other words, the potential is between the sample solution and the conductor of the pH electrode. To measure this voltage, another conductor, known as a reference electrode (whose potential does not depend on the sample), must be placed in the solution as well.
When a glass electrode is placed into an aqueous solution, the surface of the glass must come to equilibrium with the solution. Several layers of water molecules form a hydration layer on the glass membrane. The silicate structure of the glass is important as it provides locations for other cations, Na, Ca, Li, etc to reside. These may move on and off the glass surface as equilibrium is reached. Some of the Na, Ca or Li ions present in the glass might come off and be replaced by hydrogen ions in from the bulk solution. There is no evidence that any particular anions are important for pH membrane function, as fluoride, chloride, hydroxide and borate have no significant effect on the behavior. Until this hydration layer is formed and stable, the potential generated by the electrode won't be stable or reproducible. This is why it is bad for a glass electrode to dry out and for the pH-sensitive portion to be wiped with a tissue or other drying agent. It is hardly ever necessary to dry the bulb, but if it is, gentle blotting with a tissue works well. Electrodes which appear to have dried out may often be revived by soaking for a day or two in water or buffer.
Every glass membrane has an imperfection in its potential-generating nature called the asymmetry potential. A perfect glass membrane would generate zero potential when both sides are in equilibrium with solutions of the same composition. No actual membrane has this feature, or symmetry, in its physical makeup. Variations in glass composition or stresses introduced in the heating or cooling processes cause the differences between the two sides, which may reach 1 mv or more. Several of the properties relating to hydration and cation exchange capacity can be affected by the physical history of the glass bulb. Electrode manufacturers do the best practical job and know that the calibration will take care of the rest of it.
A glass electrode has another kind of error called alkaline error or sodium error. This error appears above pH 9.5 or so at room temperature in older types of glass such as Corning 015. The problem is that a sizable fraction of the Na ions are counted as protons, reducing the apparent pH even though it is a fairly basic solution. The more basic the solution becomes, the more Na ions are present and the larger the error becomes. This effect is also exaggerated by increasing temperature above 30 °C and shows a great deal of time dependence. The reason for this effect is that cation-compatible sites in the silicate structure of the glass can hold sodium or any ions smaller, including protons. The same sites that sense a proton attaching to the surface also sense the positively charged Na ion. The remedy for this behavior is to make the glass without sodium, but with the less-common and physically smaller lithium ion. The cationic exchange sites are then large enough for Li and hydrogen ions but nothing larger, including sodium. This greatly reduces the response of the glass to sodium ions. Glass electrodes described by manufacturers or catalogs as full-range (0-14 pH) use this principle in the glass composition. Other electrodes that specify a narrower range probably have a greater sensitivity to this effect, so carefully choose the electrodes for the specific application.
The behavior that leads to alkaline error can also be exploited to measure Na ions or a number of other cations and anions depending on the composition of the glass.
On the topic of glass composition, qualities that contribute to great pH sensitivity, often are flaws with regard to physical strength or chemical resistance. Each electrode is a compromise between good pH behavior and sturdiness to get the electrode lifetime up to a year or two. While cases may arise in which an electrode may last for several years, it should not be expected for an electrode to survive more than a couple of years and even less if used under harsh conditions, such as high temperature or high pH. Plan to replace electrodes every couple of years and consider devising some criterion for when an electrode is no longer meeting its specifications.
There is also an acid error, but it is more variable and harder to assign to specific physical causes. It has very little temperature dependence but does depend on the pH, time of exposure to the conditions and the anions present in the system. As acid concentration increases, the activity of the water present on the glass surface decreases, and acid error mimics the effects of less water of hydration on the surface. When placed in concentrated acid or dehydrating organic solvents, the glass electrode will give erroneously low pH readings and this is often attributed to the change in the activity of the water of hydration, but this has not been established conclusively.
Another way that pH electrodes can give erroneous results is under the influence of light. If an electrode with Ag/AgCl internal components is subjected to irradiation by UV or visible light below 470 nm, the potential generated by the electrode will be altered. Silver metal exhibits a photoelectric effect in which a potential is generated under irradiation. Silver chloride undergoes irreversible changes to silver metal on exposure to light. Studies of this effect on actual electrodes indicate that 0.2 mV/min is subtracted from the output of the electrode under strong sunlight irradiation. When the irradiation is stopped the electrode slowly recovers and behaves normally. While the magnitude of the error is variable, it can certainly amount to several tens of millivolts of error in the potential. This is of particular importance in photochemical procedures where pH needs monitoring or perhaps in an outdoor situation where a pH electrode may be exposed to strong sunlight. The proper shielding from light is needed or other internal materials should be chosen. Some electrodes have light shielding built in to greater or lesser degrees.
The Reference Electrode
In its simplest form is just a wire immersed in the same solution as the pH electrode. This arrangement, however, allows the sample to dictate the reference potential. The solution is to immerse the wire completing the circuit in its own conducting electrolyte solution that is of fixed composition and then allow a little of the solution to leak out in to the sample to complete the electrical circuit. This is basically the design of a separate reference electrode. The main design feature to select is the type of junction, meaning the junction of the internal solution with the sample. The internal conductor and electrolyte solution may also be appropriate or inappropriate for specific applications, so it pays to understand why this is so.
The most common type of junction seen today is some kind of porous frit or fiber bundle in that allows the electrolyte solution to flow out into the sample. This low to medium rate flow will often provide low resistance and a low junction potential. This flow is miniscule and is often considered non-contaminating, since, in most cases the sample is discarded and a tiny amount of KCl solution won't really change the sample. There are cases, however, in which this is not entirely true. Consider a sample containing AgNO3. A little bit of chloride leaking into that solution will instigate the precipitation of AgCl which will be evident by cloudiness. Also, certain protein samples are notorious for being unstable in the presence of the leaked KCl. Either of these precipitation situations may lead to plugging and clogging of the liquid junction. The remedy for these cases is a double junction design which has the KCl solution surrounding the Ag/AgCl internals but whose liquid junction does not flow into the sample, but into another chamber which contains a solution that is compatible with the sample. A second liquid junction allows the flow of this filling solution into the sample to complete the circuit. The result is a reference electrode which has the normal potential of the Ag/AgCl-KCl system, plus (or minus) a bit of additional potential of the second junction and electrolyte, that is compensated for by calibration.
This low-flow type of liquid junction has become standard for routine laboratory pH measurements, but there are plenty of cases which demand modifications or intentionally high flow situations, such as slurries and viscous materials. Tris buffer (short for tris-(hydroxymethyl)aminomethane, a buffer often used in biochemical applications) also requires a higher flow junction. Other junction kinds include sleeve, annular ceramic, porous junction, ceramic plug, fiber junction, double junction as well as gel electrolyte that don't actually flow out. In principle, any of these should work for "normal" more dilute, low viscosity liquid solutions. Complications arise when the wrong kind of junction is used for slurries, protein solutions, viscous liquids, very concentrated or very dilute solutions. Sleeve junctions are often the only kind used successfully in slurries found in pulp and paper applications. Low ionic strength situations such as in environmental water monitoring or waste water generally require an annular ceramic junction.
Also, consider the outward flow of electrolyte solution as a time dependent process. If the junction potential varies with the flow rate and the flow is gravity dependent, the flow will vary with the physical height of the top of the electrolyte solution. A cycle might be visible in the measured pH as the filling solution gets low, slows down, gets refilled and abruptly speeds up. Having a gas-pressured electrolye flow can remove variations that might otherwise be misinterpreted.
The outflow of the filling solution requires that the user be attentive to the level of the filling solution. The faster it flows and is depleted the more often it will need to be refilled. Different electrodes have different filling solutions and these differences should be observed. They are typically 3.0, 3.5 or 4.0 M KCl with or without AgCl. Adding AgCl helps the solution to reach equilibrium with AgCl internals faster. Calomel reference electrodes do not need AgCl in the filling solution. It is best to stay with the same electrolyte solution composition unless a good reason exists to deviate or change solutions.
A study of randomly selected pH electrodes in 1980 (J. A. Illingworth Biochem. J. 1981, 195, 259-262.) showed that most were suffering from some kind of reference junction flow anomaly that introduced substantial liquid junction potentials. These electrodes were in use in labs and presumed to be working well. He found that 24 out of 30 combination electrodes that had the porous ceramic plug type of liquid junction had significant errors. These errors were not evident in the normal two-point calibration procedure with buffers because of the similar, low ionic strengths of the buffers but were revealed by testing with higher ionic strength samples.
Another choice that can be made is the use of Ag/AgCl internals or calomel, a trivial name for mercurous chloride. There is generally no reason to use calomel which has a disposal hazard issue and it has a more restrictive temperature range, up to only 60°C or so. One reason might be the light sensitivity that Ag/AgCl shows if a photochemical irradiation or outdoor use is involved.
It is also possible to devise and to use other filling solutions, to be more chemically compatible with the samples or to introduce beneficial properties into the reference cell system. For example, when measuring samples high in organic solvent content, LiCl will work better in many situations since LiCl is more soluble in organic solvents than is the normal KCl.
Combination Electrodes
Most pH electrodes in use today are combination electrodes. That is, the pH-sensing electrode and the reference electrode are designed and constructed into the same body, often along with a temperature-sensing element. This simplifies the measurement by having only one electrode to manipulate and rinse. It can also confuse maintenance needs to be performed and precautions that accompany each type of pH glass or liquid junction type in the reference electrode. The person responsible for maintenance and upkeep should understand the pH range of the electrode and the details of the reference electrode so that the longest possible error-free use can be made of each electrode.
The pH Meter
The electronics inside the pH meter do not need a lot of discussion here but it is worth mentioning how challenging it is to measure the voltage generated by a pH electrode. A glass electrode is basically a battery whose voltage varies depending on the pH of the solution it is sensing. While the voltage generated is a few tenths of a volt, the current is very low due to the very high resistance of the glass bulb as a circuit component. The demanding part is the high input impedance required in the amplifier circuitry. Usually about 1013 ohms in most meters, this needs to be as high as practical to minimize the measurement error when the glass itself has a resistance of 106 to 109 ohms. Measuring voltage with small current flows makes the circuitry rather sensitive to temperature so that special precautions are taken in the design and component selection for the meter. It needs to generate a response to changes in the electrode and not the environment of the circuitry itself. Also, small currents can be induced by radio frequency fields from fluorescent lighting, AC motors and many electronic lab devices and as well as stray 60 Hz AC fields, noise and hum that exist around electrical equipment. Special shielding is used to protect the measurement components and amplifiers from these sources, but they can still be susceptible and users should be aware of this. Also, good connections among the components are essential. Lab environments can be corrosive and hard on metals. Corroded and oxidized contacts between the pH electrode, temperature probe and the meter will certainly hinder good measurements from taking place and the user is in a position to check on this factor.
Another point that needs emphasis is the limited role of the temperature compensation circuitry built in to most meters. The job of the meter is to match itself to the electrodes being used so that standard solutions can calibrate the meter to behave like any other "standard" pH meter. Part of this matching takes into account the temperature of the sample and electrode, which are assumed to be the same. The conversion from potential to pH is dictated by the Nernst equation, which is temperature-dependent. Whether the automatic or manual temperature compensation is used, it only reflects the change in the response of the electrode to temperature. It does not and cannot say anything about the temperature dependence of the activity in the sample. The temperature compensation circuitry may reflect the changes properly when the sample is very similar (in concentration and ionic species) to the calibration buffers, but at a different temperature. The meter can't predict the hydrogen ion activity of a sample that is very different in concentration and ionic nature, when the sample is at a temperature other than the calibration temperature. The calibration should be performed with buffers and under conditions as much like the sample as possible.
Calibration
Since electrodes cannot be manufactured with identical characteristics and are prone to change over time anyway, they need to be checked frequently and calibrated regularly against standard solutions called buffers. Calibration is actually a slight adjustment of the meter characteristics to match the actual response of the electrode system to a non-existent ideal pH meter.
For example, the first step is often to calibrate at pH 7.00 with a buffer that everyone agrees is 7.00 at 25°C. If a different calibration is being used, then the meter needs to know that different temperature and the agreed upon pH for that buffer at that new temperature. This can usually be obtained from the label on the buffer bottle or from a published chart of standard buffers, whose temperature characteristics are thoroughly known. The second buffer is often at pH 4 or 10 and determines the slope of the line defined by the Nernst equation. This control is often called slope or span on older instruments. Many modern meters identify the buffers from an internal list and so autocalibrate by recognizing two buffers from this list.
If you know your samples are always going to be in the range of 8-9 you would choose the 7 and 10 buffers to bracket your expected sample pH values. If you knew the samples will all be between 1 and 3 you might choose a pH 1 buffer along with pH 7 or even choose 1 and 4 as the calibration points if you expect few samples outside this range.
Certain pH ranges at the extremes of the scale are especially sensitive to temperature. If you know you are working at a pH extreme you will want to pay special attention to matching the temperature of the calibration buffers to that of the samples.
Electrode Precautions and Electrode Maintenance
Dry pH electrodes need to be soaked in deionized water or buffer for at least 24 hours before potentials will stabilize.
Don't let a glass electrode go dry.
Don't place a glass electrode into a dehydrating solution such as ethanol, sulfuric acid, etc.
Make sure that pH electrode is compatible with pH extremes if measurements will be commonly performed at the ends of the pH scale. That is the glass type must allow good measurements and the product literature or catalog will give an acceptable pH range for each model of electrode.
Electrodes should be stored in pH 4 or 7 buffer. Alkaline buffers will dissolve the glass at a much greater rate and greatly reduce the lifetime of the electrode.
Reference electrodes should have their flow verified or the junctions cleaned regularly. It helps to replace the filling solution once a month or so. Some deionized water can be forced through the junction with a little bit of air pressure from a pipet bulb. This will demonstrate that the junction is not clogged and dissolve any crystals that may be forming.
The reference electrode section should be filled to nearly the top or the filling hole with the proper filling solution. Pay attention to the proper filling solution and don't mix up 3M, 3.5M, 4M and saturated KCl solutions and all these are available with and without AgCl. Silver/AgCl electrodes use AgCl in the filling solution while calomel electrodes do not. Know your reference electrode types and what solution they require.
Plugs, stoppers, caps and screw-type closures on reference electrodes should be kept closed when not in use to preserve filling solution. They should definitely be open during measurements though or the minimal junction potential won't be achieved.
Reference and pH electrodes should be inspected for air bubbles that can interfere with establishing a stable potential. Careful but vigorous shaking can dislodge the bubbles.
If a stable reading can't be established on the calibration buffers in 1 minute or less then the electrode may be in poor condition and may need replacing.
Recording the zero point and slope in a logbook at each calibration can show when a deviation has occurred and attention is needed.
Make sure that buffers are not used past their expiration date.
Calibrations should always be performed in a similar manner, with the same stirring speed, temperature and at the same temperature as the unknown samples.
When measuring a suspension, three values are possible depending on whether the probe is in the well-stirred slurry, in the supernatant liquid above the solids or sitting in the solids on the bottom of an unstirred vessel. A rational for choosing one of these situations should be developed and then performed the same way each time.
After rinsing electrodes, they should generally not be wiped as this causes several difficulties. A static charge can be transferred to the glass bulb which then may take some time to discharge, interfering with stable potentials. Wiping can also disturb the hydration layers which also take time to re-establish their equilibrium and give stable potentials.
Glass electrodes used in non-aqueous situations need frequent re-hydration for the pH function to be active.
Beware of calibration at lower temperatures as the glass resistance increases as the temperature goes down. Chemically, the kinetic processes involved in equilibration take longer at lower temperatures. Electrically, higher resistance in the glass and the sample can result in a noisy signal. It's possible, just harder.
Bibliography
Atkins, P.W. Physical Chemistry W.H. Freeman: San Francisco, 1978.
Castellan, G.W. Physical Chemistry 2nd ed. Addison-Wesley: Reading, MA, 1971.
Perin, D.D.; Dempsey, B. Buffers for pH and Metal Ion Control. Chapman and Hall: London, 1974.
Wescott, C.C. pH Measurements. Academic Press: Orlando, FL, 1978.
Bates, R.G. Determination of pH: Theory and Practice. Wiley, New York: 1973.
