What Is a Reference Standard? The Ruler a Lab Measures Against
A reference standard is a sample whose identity and amount are already trusted, kept so that other samples can be measured against it. It is the metre stick in the vault that every tape measure is ultimately checked against.
A reference standard is a sample whose identity and amount have already been established, kept so that unknown samples can be compared with it. Without one, a laboratory measurement is only a reading from an instrument. With one, the reading can be turned into a statement such as this is the substance we think it is, and this much of it is present.
That raises a fair question about a figure such as 99% pure on a certificate: compared with what? The answer turns out to be partly nothing, and partly a standard. This article explains the idea from zero, the difference between primary and secondary standards, how a standard turns a reading into a number, and what happens when a new compound has no standard at all.
A known thing to compare the unknown against
Start with a ruler. For a long stretch of the nineteenth and twentieth centuries the length of a metre was defined by a particular metal bar, kept under controlled conditions near Paris. Copies were made and compared with it, then copies of the copies went out to workshops. Every tape measure in the world could trust itself only because, a few steps up the chain, someone had checked it against that bar. The definition of the metre has since changed, but the lesson about chains of comparison has not.
A reference standard does the same job in a chemistry laboratory. It is a sample of a substance whose identity and quality have been established with enough care that other measurements can lean on it. A laboratory that wants to know whether an unknown sample is a particular compound, or how much of that compound it holds, runs the standard and the unknown side by side and compares the two results.
The reason is simple. Analytical instruments do not announce the truth; they produce signals, and a signal means something only when it is compared with the signal from a material of known identity and content. Suppliers of laboratory materials explain the need for standards in much the same terms 3. The standard is what gives the signal a meaning.
Primary, secondary and working standards
Not every standard sits at the same level of the chain. The international guide for making the active ingredients used in medicines defines two main kinds 1.
A primary reference standard is a substance shown by an extensive set of analytical tests to be authentic material of high purity. It can be obtained from an officially recognised source, prepared by independent synthesis, taken from existing high-purity production material, or prepared by further purifying existing production material 1. A secondary reference standard is a substance of established quality and purity, shown by comparison with a primary standard, and used as the reference for routine laboratory analysis 1.
In everyday laboratory language the second kind is often called a working standard. It is the one in the cupboard that gets used week after week, because the primary standard is precious and expensive. The secondary is calibrated against the primary, which is how the chain of trust is kept intact without using up the original 2.
| Level | What it is | How its value is established | Typical use |
|---|---|---|---|
| Primary standard | High-purity material, authenticated by extensive testing | From an officially recognised source, or fully characterised by independent methods | Setting or checking other standards |
| Secondary standard | Material of established quality | By comparison with a primary standard | Routine analysis |
| Working standard | A secondary standard in everyday use, often in-house | Requalified against a primary or secondary standard on a schedule | Daily runs; replaced when it expires |
How a standard turns an instrument reading into a number
An instrument reports a signal. A standard translates it. Take the technique described in the article on HPLC, in which a sample is separated and each substance produces a peak at a particular time. The size of the peak depends on how much of the substance went through, but the instrument has no idea what size corresponds to how much.
So the laboratory first runs the standard, whose amount is known. Suppose, as a pure toy example, the standard produces a peak of a certain size. If the unknown sample, prepared in the same way, produces a peak half that size, then it holds about half as much. The standard has turned a size on a chart into an amount. It also gives a second thing: the time at which the standard's peak appears. A peak in the unknown at the same time is consistent with the same substance being present.
Notice what the standard is doing in each case. It is not changing the instrument or making it more accurate. It is providing the one thing the instrument cannot supply itself: a known point of comparison.
Certified reference materials and who certifies them
Some standards come with a formal certificate. A certified reference material is one issued with documentation stating its assigned value, the uncertainty around that value, and how the value traces back to a higher authority. The top of the chain is a national or international standards body 2. Uncertainty here is not a confession of error; it is an honest statement of how precisely the value is known.
Who issues them? Several kinds of body do. National measurement institutes produce certified materials for many chemicals. Pharmacopoeial organisations, the groups behind the official books of quality standards, supply reference standards that go with their published tests, so that a laboratory running a monograph test has the right material to compare against 4. Specialist producers also supply them, and an international standard sets out the competence a reference-material producer should demonstrate.
Storage and age matter too. A standard has a validity period, and results obtained with an expired one carry no credibility until a fresh standard has been prepared and requalified 2. The trust is in the chain and its dates, not just in the jar.
Why a purity figure without a standard is weaker than it looks
Back to 99% pure. The usual meaning on a certificate is a share of peak area: of everything the detector saw, ninety-nine parts in a hundred came from the main peak. Working that figure out needs no reference standard at all. It compares the sample with itself.
That is useful, but it answers a narrow question: how much of what the instrument could see belongs to the biggest peak. It cannot say whether the biggest peak is the compound named on the label, because it has nothing outside the sample to compare with. Identity needs a comparison, and so does the actual amount of material in the container. Anything the detector cannot see does not enter the sum at all.
This is why a purity figure and a reference standard answer different questions. A high share of peak area with no standard behind it tells you the sample is simple. It does not tell you what it is made of.
What happens when no standard exists for a new compound
A standard has to come from somewhere. For long-established substances, an officially recognised source exists. For a new research compound there often is none: nobody has yet made, characterised and certified a batch to serve as the world's yardstick.
The manufacturing guide anticipates this. Where a primary standard is not available from an officially recognised source, it says an in-house primary standard should be established, with appropriate testing to establish its identity and purity fully, and with the testing documented 1. In other words, the laboratory builds its own ruler and shows its working.
That is a legitimate approach, but it changes what the result means. A measurement against an in-house standard is only as good as the characterisation of that standard, and no outside body has agreed to it. It is not wrong. It is a claim that rests on a document the reader should be able to ask about, rather than on a shared yardstick that the whole field already accepts.
Questions a beginner can ask about a certificate
None of this requires a chemistry degree. It requires a few plain questions, asked of the document rather than of the person who issued it.
- Does the certificate say what the result was compared with, or only give a figure?
- If a standard was used, is it named, and is it an official one or an in-house one?
- Is the standard identified by lot and date, so that it can be traced?
- Does the figure refer to share of peak area, or to the actual amount of the compound? These are different things.
- Is the method named, and could another laboratory repeat it?
A certificate that answers these has told you how its number was made. One that does not has still told you something: that the number comes with no visible chain behind it. Neither case is a verdict on any product. They are simply the difference between a figure you can check and a figure you can only repeat.
Nothing here is guidance about using any substance. Material sold for laboratory research use only has not been assessed by a medicines regulator, and a reference standard is a tool for measuring, not a statement that anything is safe or suitable for any person.