Conductivity measurement: cells, temperature and applications

A meaningful conductivity result needs more than a stable number: the cell must suit the sample and range, the wetted geometry must be controlled, and the temperature basis must be known. Before comparing two results, check whether they represent the sample at its actual temperature or a value calculated for a reference temperature. Those are different reported quantities.

Conductivity is sensitive to mobile ions; it does not identify them or establish overall chemical purity. This guide develops a bench measurement workflow, a cell-selection matrix and a simulated temperature example. It addresses routine operation, not pharmaceutical water-system design or a complete calibration procedure. Acceptance decisions must use the applicable method, edition and laboratory procedure.

What conductivity tells you—and what it cannot establish

In an electrolyte, charge is carried by moving ions. The result depends on the species present, their concentrations and mobilities, and the medium and temperature. Two chemically different samples can therefore have similar conductivity. Conversely, changing temperature or composition can change the reading without any fault in the meter. Interpret the measurement in relation to a defined sample and analytical question.

Conductance, G, describes the electrical response of the particular cell and sample arrangement and is expressed in siemens, S. Conductivity, κ, relates that response to the cell constant, K: κ = K × G. When K is expressed in cm⁻¹ and G in S, κ is in S/cm. Resistance and resistivity are likewise different quantities; recording an unlabelled number obscures what was actually measured.

A conductivity check can detect an ionic change relevant to a process or material, but it cannot by itself prove identity, quantify every contaminant or exclude non-ionic contamination. A low value is not a general purity certificate. GuideGxP recommends defining the intended decision first: screening an incoming solution, following a preparation step, or performing a specified compendial test may require different controls.

Choose the cell for the sample and working range

Cell constant reflects electrode geometry, including spacing and effective area. Real measurement behaviour also depends on the cell design and operating range; a nominal K is not evidence that every solution can be measured correctly. Check the documented conductivity range of the complete meter–cell combination, its materials, available sample volume and the compatible temperature sensor before selecting it.

Two-electrode and four-electrode arrangements address measurement constraints differently. Separating current-carrying and voltage-sensing electrodes can reduce certain electrode and lead effects, but four electrodes do not automatically guarantee better results for every low-conductivity sample. The arrangement, electronics, range and verified performance must be considered together. Do not infer a universal “best” cell constant from a single application.

The matrix below is an editorial decision aid, not a specification or validated method. A suitable choice requires evidence for the actual matrix and intended range. Cleaning compatibility, accessible wetted surfaces and safe handling are part of that choice; a geometrically suitable cell can still be unsuitable for a coating, solvent or aggressive sample.

Temperature: distinguish measurement from compensation

Measure temperature where it represents the liquid interacting with the cell. A probe in another part of the vessel, a warm sample entering a cool cell or a rapidly changing sample can create a mismatch even when both displays appear plausible. The measurement condition should specify how equilibration and stability are assessed; there is no universal waiting time that proves suitability for every matrix and volume.

Temperature compensation calculates an estimated conductivity at a reference temperature using a defined relationship. It does not physically bring the sample to that temperature. The appropriate relationship can depend on composition and range; a coefficient suitable for one solution need not describe another. “Automatic” identifies an implemented function, not evidence that the model fits the sample.

The applicable method determines whether to report compensated or uncompensated conductivity. Do not impose compensation on every compendial test. USP ⟨644⟩ distinguishes its scope from ⟨645⟩ for water; the public introductory previews are not a substitute for the current complete chapters. Confirm the relevant instructions before configuring a release measurement.

A simulated example: one signal, three reporting settings

Assume a hypothetical measured conductivity κ_T of 1.20 mS/cm at 35 °C. For this illustration only, use a linear model with a reference temperature of 25 °C and an assumed coefficient α = 0.020 °C⁻¹. This coefficient is invented for the calculation; it is neither an experimental finding nor a recommended default.

κ_ref = κ_T / [1 + α(T − T_ref)]. The denominator is 1 + 0.020 × (35 − 25) = 1.20, giving 1.00 mS/cm at the modelled reference temperature. With a different assumed coefficient, 0.015 °C⁻¹, the same input gives 1.20 / 1.15 = 1.043478… mS/cm, approximately 1.04 mS/cm.

The uncompensated value 1.20 at 35 °C and the calculated values 1.00 or 1.04 referenced to 25 °C are not interchangeable results. Their difference alone does not prove drift, contamination or a failing cell. First inspect the reporting basis. Never choose a coefficient because it makes a result comply; the model and its applicability must be established independently of the desired outcome.

Control immersion, bubbles and sample handling

Follow the approved cell geometry: required electrodes, openings and the relevant temperature-sensing region must contact the sample as intended. Keep the specified relationship to vessel walls and the bottom. Moving the cell between dissimilar containers or reducing volume may change its effective environment. Confirm suitability rather than assuming that any visible immersion is sufficient.

Inspect for bubbles on sensing surfaces and in cavities. Remove them only by a handling action compatible with the cell, sample and procedure; vigorous shaking can introduce gas, change temperature or alter a sensitive sample. If bubbles recur, investigate the filling or immersion technique. Do not dismantle a sealed cell or alter its electrical connections to diagnose the problem.

Control carryover between standards, concentrated samples and dilute samples. Use a documented rinse and, where necessary, conditioning sequence appropriate to the material and matrix. Residual rinse liquid can dilute a small sample; residues on a cell can increase a low-conductivity reading. Neither one rinse volume nor one cleaning chemistry is universally suitable. Avoid returning measured aliquots to the original sample container.

Record sample identity, preparation, vessel, volume where relevant, exposure and mixing conditions. A closed arrangement can reduce atmospheric interaction when the method supports it, but it is not automatically equivalent to an open beaker measurement. Do not strip dissolved gases or otherwise modify the sample merely to lower the result. Handle hazardous solvents, hot liquids and electrical equipment under the relevant laboratory controls.

Low and high conductivity: different vulnerabilities

At low conductivity, a small ionic contribution from a previous sample, a vessel or environmental exposure can be significant relative to the sample itself. Carbon dioxide absorption can change aqueous composition and conductivity. A rising reading therefore needs an investigation of sampling and exposure as well as the instrument. A blank helps only when its preparation and interpretation are appropriate to the analytical question.

At high conductivity, verify that the full system remains within its demonstrated range. Electrode effects, configuration and instrument limitations may matter. An out-of-range indication is not a valid numerical result, and changing the range setting cannot expand a cell’s validated capability. Dilution changes the sample; use it only when the method permits it and its effect on interpretation has been established.

Resistivity is the reciprocal of conductivity when units and conditions are consistent. It adds no independent chemical identification. The example is a mathematical conversion, not a pharmaceutical water acceptance value. Keep actual or reference temperature and compensation status attached to the converted result; taking the reciprocal cannot remove uncertainty or sample-handling effects.

A total dissolved solids display usually applies a relationship to conductivity. A universal TDS factor does not exist for all mixtures, and non-ionic constituents need not be represented by an ionic response. Do not relabel conductivity as a measured mass concentration without a justified relationship for the sample and intended use. Confirm which quantity the instrument stores and exports.

Investigate unstable, unexpected or out-of-range readings

Start by preserving the original result, flags and settings. Check sample identity, units, selected cell/channel, entered or stored cell constant, temperature reading, compensation model and reference temperature. A change from µS/cm to mS/cm or a different reporting basis can explain an apparent disagreement without any physical change in the sample.

Next inspect immersion, bubbles, contamination, connectors and the temperature probe without opening energized equipment. A damaged cable, incompatible sensor or persistent temperature fault requires appropriate technical assessment. Do not bypass interlocks, modify circuitry or improvise an electrical test. Remove unsuitable equipment from use under the laboratory procedure when necessary.

Plan targeted checks that distinguish hypotheses. A suitable control behaving normally while a low-conductivity sample rises during exposure points toward sample-handling questions, but does not alone prove a specific contaminant. If both control and sample are inconsistent, investigate the shared cell, settings, temperature and preparation. Change one justified condition at a time and record the reason and outcome.

A controlled comparison with another suitable, verified system may help, provided both systems measure equivalent aliquots under comparable conditions. Repeated measurements without a defined question can conceal sample evolution. Follow the applicable investigation procedure for suspect or out-of-specification results; do not discard early readings, average away a trend or adjust settings until a passing number appears.

Routine controls and the metrological interface

Keep a record of cell and meter identity, measurement range, method version, maintenance and relevant checks. The analytical record should retain conductivity, units, actual temperature, compensation status, model or coefficient, reference temperature, stability assessment and any abnormal observation. Where software stores only a converted value, assess whether the retained metadata support reconstruction of the reported result.

Calibration establishes a metrological relationship; adjustment changes the measuring system, and verification checks specified requirements. These activities are related but distinct. Instrument qualification, method suitability and maintenance answer further questions. A calibration record does not automatically demonstrate that a viscous or mixed-solvent sample can be measured meaningfully with the chosen geometry.

Define check materials, limits and frequency through the approved method, intended use, risk and performance history. Do not borrow an arbitrary universal interval or tolerance. After cleaning, replacement or a consequential adjustment, establish which checks are required before returning the system to use. Document unsuccessful controls and their disposition as well as acceptable ones.

Detailed conductivity calibration belongs in the separate laboratory metrology workstream; this article provides only the operating interface. The practical conclusion is to compare results only after aligning sample handling, cell suitability, units and temperature basis. If those conditions are undocumented, resolve that gap before attributing a difference to sample quality or instrument failure.

Sources and applicability

NIST SP 260-142 is a historical metrological reference used for physical principles, not a current pharmaceutical acceptance standard. The public USP ⟨644⟩ archive was read in full but carries a 2019 official date; it is not evidence of the complete current text. The linked USP ⟨644⟩ and ⟨645⟩ records provide introductory previews only. Check the complete applicable edition before making a compendial decision. The matrices and investigation sequence are GuideGxP recommendations; the numerical case is simulated.

1. Shreiner RH, Pratt KW. Primary Standards and Standard Reference Materials for Electrolytic Conductivity. NIST SP 260-142, May 2004, §§1–3, 5.
2. USP–NF ⟨644⟩ Conductivity of Solutions. Official introductory preview, bibliographic record 2020.
3. USP ⟨644⟩ Conductivity of Solutions. Public harmonization archive, September 2018; official date shown: 1 August 2019.
4. USP–NF ⟨645⟩ Water Conductivity. Official introductory preview, bibliographic record 2017.
5. JCGM. International Vocabulary of Metrology, VIM, §2.39: calibration.

Related analytical laboratory topics

• Analytical instrument qualification and USP ⟨1058⟩
• Qualification, method validation and system suitability
• pH measurement: electrodes, difficult samples and troubleshooting