Glossary of Terms

The degree of statistical agreement (usually at the 95% confidence level) between a measured value and the true value, or, the certainty or sureness with which a measured value is known. May be reported in relative or absolute units. Absolute accuracy is expressed in the same units as the concentration of the component. Relative accuracy is expressed as a percentage of the concentration of the component. Example: 4.95% – 5.05% expressed in absolute accuracy terms is 5% ±0.05% absolute. It can also be expresed as 5% ±0.05/5 in relative terms, or ±1% relative.

A process based on statistics designed to estimate the true concentration for each component of a mixture and to assign the analyzed concentration. A further objective of analysis is to determine the analysis accuracy for each component. Analytical measurements have sufficient work content to assure that component concentrations and accuracies meet customer and Scott specifications.

The degree of statistical agreement between the analyzed and the true concentration for a component of a gas mixture (usually at the 95% confidence level), or, the certainty or sureness with which the analyzed concentration is known. Analytical measurements have sufficient anchoring to reference standards, calibration, precision, resolution and traceability to assure that the true concentration for a component can be estimated to the claimed accuracy. Analytical accuracy can be calculated, because Scott analytical measurements require actual data for reference standards, calibration, precision, resolution and traceability.

The degree of agreement (usually the simple ratio) between the analyzed concentration and the requested concentration for a component. Analytical tolerance is not determined by statistical means (whereas accuracy is). Analytical tolerance is sometimes used synonymously (but incorrectly) with analysis accuracy.

The numerical assigned to a component based on the results of analysis. An analyzed concentration is expressed in both concentration units (i.e. percent, parts per million) and concentration basis (i.e. mole/mole, weight/weight). An analyzed concentration is also accompanied by an analysis accuracy value (i.e. ±2%).

The material (component) making up the majority of a mixture. The balance may be either gas or liquid phase. The balance may also be a mixture (i.e. air). If the mixture is entirely gas phase, the balance is referred to as the balance gas.

A systematic error in a process which causes all measured values to deviate from the true value by a consistent amount in a consistent direction (higher or lower than).

The degree of statistical agreement between the blended and the true concentration for a component of a gas mixture (usually at the 95% confidence level), or the certainty or sureness with which the blended concentration is known. Blended accuracy can be calculated, because Scott analytical measurements require actual data for reference standards, calibration, precision, resolution and traceability.

The degree of agreement (usually the simple ratio) between the blended concentration and the requested concentration for a component. Blend tolerance is not determined by statistical means (whereas accuracy is).

The numerical concentration assigned to a component based on the results of analysis. An analyzed concentration is expressed in both concentration units (i.e. percent, parts per million) and concentration basis (i.e. mole/mole, weight/weight). A blended concentration may also be accompanied by a blend accuracy value (i.e. ±2%).

A process by which component and balance gases of a gas mixture are metered into a container. Blending process measurements (i.e. scale resolution, actual weight data, partial pressures, Acublend® response data) have sufficient resolution and accuracy to assure that components are in the required ratio to meet customer and Scott specifications.

A process designed to determine the relationship between detector response and concentration for a given component on an instrument. During calibration, one or more samples of known concentration are introduced to the detector and detector response is recorded for each sample. Calibration may be performed with reference standards, or with calibration standards accompanied by reference standard verification.

The mathematical function produced by regression of the detector responses recorded during calibration of an instrument. The function describes detector responses over a range of concentrations and is used to predict the concentration of an unknown sample based on its detector response.

A range of concentrations between the highest and lowest concentrations of standards used for calibration, from which the concentrations of unknown samples can be determined. Concentration range may be further limited by precision and accuracy requirements.

A document which represents the certified concentration of a component of a mixture to the customer. Scott Certificates of Analysis require actual analysis and backup data to substantiate concentration and accuracy claims.

The numerical concentration assigned to a component that is printed on the certificate of analysis or conformance. This is the concentration represented to the customer for that component and the concentration that the customer is supposed to use. The certified concentration may be either the blended or analyzed concentration.

cGMPs are the Current Good Manufacturing Practices followed by the pharmaceutical and device firms to ensure that the products produced meet specific requirements for specification, identity, strength, quality and purity. FDA regulates these industries to ensure cGMPs are being followed by using the regulations in 21 CFR and various guidance documents.

A constituent of a mixture, either a minor component or balance. A two-component mixture consists of one minor component and a balance material. There is no such thing as a one-component mixture; this is referred to as a pure material.

The ratio or proportion of the quantity of a given component to the total quantity of mixture. A concentration is expressed in terms of concentration units (i.e. percent, parts per million) and concentration basis (i.e. mole/mole, weight/weight).

The basis (the counting units) for calculating the quantitative ratio of a component to the total mixture (i.e. mole/mole, weight/weight). Example: A Certificate of Accuracy stating “Component X: 100 ppm ±2% (mole/mole)”, (i.e. analyzed concentration ± analysis accuracy @ concentration basis), means “for every 1,000,000 total moles of mixture contained within this cylinder, 100 moles (no less than 98, no more than 102, at the 95% confidence level) are Component X.”

An undesired component in a pure gas or mixture, either known or not known to be present.

The length of time over which Scott will guarantee that components of a gas mixture remain at their certified concentrations within the specified tolerances, or the stated date after which time the mixture is no longer valid. An expiration date is based on Scott history, experience, and stability studies, or is arbitrarily set by regulations (i.e. EPA protocols).

Pure materials or gas mixtures in which all components remain in the vapor state under the conditions (i.e. temperature) recommended for use of the product.

The mixture preparation process in which the individual components are weighed into the cylinder using high pressure, highly sensitive electronic balances.

An additional, potentially undesirable component of a pure gas or mixture. Impurities are most commonly encountered in the pure material used as the raw material source for a component of a gas mixture. An impurity may be removed by purification or quality controlled and accounted for during blending, thereby preventing it from becoming a contaminant.

A component with no uncontrolled chemical incompatibility with other components or with contaminant materials of construction. Examples of inert components include nitrogen, helium and carbon dioxide.

Pure materials or gas mixtures in which one or more components exist in a liquid state under the conditions (i.e. temperature) recommended for use of the product. Liquid phase materials can be stored in equilibrium with the gas phase (conventional cylinder) or forced to an all-liquid state (piston cylinders).

For a given component, the matrix is the background of all other gases present in the mixture. This term is most frequently used to contrast two mixtures with a common minor component, but with different background of other components which affect blending or analysis. Matrix matching is one technique which can be used to overcome interferences caused by matrix effects. Example: Mixture A contains 500 ppm of NO in a nitrogen balance and Mixture B contains 500 ppm of NO, 20% CO2 and a nitrogen balance. The matrix for Mixture A is nitrogen, whereas the matrix for Mixture B is nitrogen with 20% CO2. The analysis for NO in these two mixtures is radically different.

A component making up a small fraction of a gas mixture, for which specifications have been requested by the customer. Usually, a minor component is a material which is blended and analyzed in a gas mixture.

A classification standards of weights that are traceable to the US National Institute of Standards and Technology (NIST). Class S signifies that the weights are for precision laboratory standards for scientific use, and are used for reference, calibration and precision analytical work.

The degree of dispersion (closeness) of repetitive, equivalent readings of measured value. Precision is one of the critical contributors to overall analysis accuracy, and must be known and controlled during analysis. Precision is commonly expressed in terms of CV% (Coefficient of Variation %) or RSD% (Relative Standard Deviation %).

The natural variation of each of the components of an analysis process, which causes measured values to deviate from each other to some degree. Repetitive measurements which consist of only random errors (no bias errors) will from a normal (Gaussian or bell-shaped) distribution around the true value.

A component with known chemical incompatibility with other components or with contaminant materials of construction. Examples of reactive components include sulfur dioxide, alcohols, hydrogen sulfide, ethylene oxide, etc.

The origin of, or the transfer vehicle for, the true value of a measured quantity. Analysis can only be performed, and the true concentration for a component determined, using a reference standard with an established tracebility. The highest level of reference standards originate form standardization agencies (NIST, NMi), and include items like SRMs, PRMs and Class S Weights. Lower level reference standards may also be permanent (retained cylinders) or temporary (flasks, permeation tubes).

The degree of agreement between the concentration of an old cylinder, used as a standard, and the concentration of a new cylinder, analyzed as an unknown. Agreement between these two cylinders essentially verifies that the calibration of the instrument used to make the comparison is still valid and no recalibration is required. The closeness of agreement is specified by the customer, depending on the calibration needs of the application. Scott controls calibration reproducibility by controlling analytical accuracy (for dual-certified products) or process accuracy (for Sure Class™ products). Calibration reproducibility is calculable once the customer has specified analytical accuracy. Conversely, if the customer requires a specific calibration reproducibility, then the required analytical accuracy can be inferred.
Calculation: Root sum square of the analytical accuracies of the two cylinders to be intercompared.
Example: The customer specified two cylinders, each with ±2% analytical accuracy. These two cylinders can be expected to intercompare with each other with 2.8% (assuming no calibration curve errors on the customer’s instrument).

The extent to which an instrument is capable of distinguishing closely adjacent responses.

The degree to which components of a gas mixture will remain at their certified concentrations within specified tolerances over time.

The difference between a measured value and a Scott or customer-specified value.

The relationship between a measured value and an established element of the National/International Measurement Systema, or the basis for establishing the true value of a standard. For traceability to exist, there must be an unbroken chain of comparisons between the sample and the National/International Measurement System. Elements of the National/International Measurement System can be fundamental units of measurement (i.e. weight, temperature, pressure) or secondary/transfer standards (i.e. synthetic gas standards).