Categories
Analytical Methods and Purity

Declared purity is a measurement result, not a property of the material. Two valid methods can give different numbers for the same sample.

The certificate says “≥99%”. The question this number answers is much narrower than is usually assumed — and until you know which method produced it, you know practically nothing. The difference between 98% and 99% is often less significant than the difference between two ways of calculating the same 99%.

This article explains what lies behind a reagent’s stated purity: where the number comes from, how chromatographic purity differs from substance content, why two laboratories can report different values for the same sample, and what is worth checking in practice before a material is accepted as fit for use.

The number is an answer to the question asked by the method

There is no single “purity”. There are results of specific determinations, and each of them measures something different. Three approaches are encountered most often, and it is worth being able to tell them apart from the description on the certificate alone.

Chromatographic purity is the share of the main peak’s area in the sum of the areas of all peaks in the chromatogram. It is a relative quantity: it tells you what fraction of the recorded signal is the main substance. It does not, however, tell you how much of the substance is in the sample by weight, because it does not see anything that did not produce a peak — and among the things that produce none are water, inorganic salts, and compounds that do not absorb at the selected wavelength.

Mass balance approaches the matter the other way around: all measurable impurities — water, residual solvents, sulfated ash, organic impurities — are determined, and their sum is subtracted from one hundred percent. This is how, for example, amino acid reference materials are certified.1 This approach gives a result closer to the actual content by weight, but it is sensitive to whether all classes of impurities were taken into account at all.

Quantitative nuclear magnetic resonance (qNMR) measures something else again: the ratio of the number of nuclei in the sample to the number of nuclei in a standard of known purity. This method has been used, among other applications, to determine the purity of pharmaceutical reference materials using proton spectra.2 Its advantage is independence from whether an impurity produces a chromatographic peak — if it contains hydrogen, it will be counted.

Why two laboratories report different numbers

In the certification of reference materials, the standard is to combine at least two independent methods. A good example is the certification of a caffeine reference material, in which purity was determined by ultraviolet spectrophotometry and verified by quantitative NMR.3 The point of such a combination is that the two methods have different blind spots: what one fails to see, the other has a chance to catch.

A discrepancy between results does not have to mean that one of the laboratories made a mistake. Most often it means that the determinations were answering different questions. A material containing one percent water will get excellent chromatographic purity and a clearly lower content by mass balance — and both numbers will be correct.

Chemical metrology has its own language for these distinctions; the general requirements for reference materials and their role in quality assurance have been described in review papers devoted to this class of materials.4 The practical conclusion for the recipient is simple: a number without the name of the method is incomplete.

The peak purity trap

The most frequently underestimated limitation of chromatography is coelution. If an impurity leaves the column at the same time as the main substance, its area will be added to the main peak — and the result will be higher than the actual one. Chromatographic purity can therefore err in the direction favorable to the material.

This is why separate procedures have been developed for assessing the purity of the peak itself, for example using principal component analysis of diode-array detector data.5 Such an examination answers the question of whether the peak is spectrally homogeneous across its entire width — or whether two substances are hiding under a single apex. Without this verification, a declaration of “99% by HPLC” rests on the tacit assumption that the separation was complete.

Purity determination methods are developing precisely in the direction of reducing this uncertainty: internal standard correction is used,6 as are variants with signal suppression,7 and also approaches that make it possible to determine many substances using a single reference standard.8 All these works share one assumption: the area under the peak alone is not enough.

How 98% differs from 99%

Arithmetically, the difference is one percentage point. Methodologically, the question is different: what that one percent is, and whether it is known what the remaining two percent in a 98-percent material is.

In analytical work, the identity of impurities usually matters more than their sum. One percent of water is a weight problem — the material has to be dried or a correction has to be applied. One percent of an isomer or a structurally close analog is a problem of an entirely different class, because such an impurity may behave similarly in most determinations and remain invisible until the moment it starts to interfere. One percent of residual solvent is yet another situation — usually easy to detect and easy to remove.

Hence the practical order of questions when assessing a material:

  • What method produced the number? HPLC with UV detection, mass balance, qNMR — these are three different answers.
  • Was the water content determined? Without it, chromatographic purity and content by weight can diverge significantly.
  • Was the homogeneity of the main peak checked? If not, the result contains an assumption of complete separation.
  • Was the identity confirmed independently? Purity tells you about proportions, not about what the main substance is.

The last point is often skipped, yet it is the most important. A purity determination takes the identity of the material as a given and calculates what fraction of the whole it represents. A 99.9% pure material with a misassigned identity is useless, and its certificate looks excellent. This is why identity confirmation — by NMR spectrum, mass spectrometry, melting point — is a separate step, not a conclusion drawn from purity.

What this looks like in pharmaceutical analysis practice

In pharmaceutical analysis, separating the assay of content from the evaluation of impurities has long been standard; it has been described, for instance, in the context of validating reversed-phase chromatographic methods for individual substances.9 Similarly, in the certification of plant-derived standards, many independent techniques are used so that the result does not rest on a single method.10

The limitations of liquid chromatography itself and how to read a certificate of analysis (CoA) — including the fields that are most often left empty — are covered in separate articles in this series. The full catalog of reagents with CAS numbers and declared purity is available in our shop.

If a single sentence were to remain from this entire article, it would be this: declared purity is a measurement result, not a property of the material. Two materials with an identical number on the certificate can differ in what nobody measured — and it is precisely this difference that usually decides whether a determination succeeds.

Sources

Each item below was retrieved from the PubMed database via the E-utilities interface and is not reproduced from memory. PMID numbers link to the source records.

  1. Liu H, Cheow PS, Yong S et al. Determination of purity values of amino acid reference materials by mass balance method: an approach to the quantification of related structure impurities. Anal Bioanal Chem. 2020;412:8023-8037. PMID: 32914399. DOI: 10.1007/s00216-020-02936-7.
  2. Monakhova YB, Kohl-Himmelseher M, Kuballa T et al. Determination of the purity of pharmaceutical reference materials by 1H NMR using the standardless PULCON methodology. J Pharm Biomed Anal. 2014;100:381-386. PMID: 25215441. DOI: 10.1016/j.jpba.2014.08.024.
  3. Shehata AB, Rizk MS, Rend EA. Certification of caffeine reference material purity by ultraviolet/visible spectrophotometry and high-performance liquid chromatography with diode-array detection as two independent analytical methods. J Food Drug Anal. 2016;24:703-715. PMID: 28911607. DOI: 10.1016/j.jfda.2016.06.009.
  4. Kiełbasa A, Gadzała-Kopciuch R, Buszewski B. Reference Materials: Significance, General Requirements, and Demand. Crit Rev Anal Chem. 2016;46:224-35. PMID: 26042643. DOI: 10.1080/10408347.2015.1045120.
  5. Wiberg K, Andersson M, Hagman A et al. Peak purity determination with principal component analysis of high-performance liquid chromatography-diode array detection data. J Chromatogr A. 2004;1029:13-20. PMID: 15032344. DOI: 10.1016/j.chroma.2003.12.052.
  6. Sun X, Zhang W, Huang T et al. Purity determination of pyributicarb by internal standard correction-high-performance liquid chromatography-quantitative nuclear magnetic resonance. Anal Bioanal Chem. 2020;412:6983-6993. PMID: 32754793. DOI: 10.1007/s00216-020-02832-0.
  7. Wu H, Huang T, Zhang W et al. Purity determination of digitoxin by two-signal suppression-internal standard correction-high-performance liquid chromatography-quantitative nuclear magnetic resonance. Anal Bioanal Chem. 2025;417:2935-2946. PMID: 40119928. DOI: 10.1007/s00216-025-05836-w.
  8. Kitamaki Y, Saito N, Yamazaki T et al. Determination of PAHs in Solution with a Single Reference Standard by a Combination of (1)H Quantitative NMR Spectroscopy and Chromatography. Anal Chem. 2017;89:6963-6968. PMID: 28581717. DOI: 10.1021/acs.analchem.6b05074.
  9. Colgan ST, Dumont ML, Ruggeri SG. Assay and purity evaluation of sunepitron hydrochloride by reversed-phase liquid chromatography using a reference standard composite. J Pharm Biomed Anal. 1998;18:429-40. PMID: 10096837. DOI: 10.1016/s0731-7085(98)00057-0.
  10. Song JZ, Qiao CF, Li SL et al. Purity determination of yunaconitine reference standard using HPLC with experimental design and response surface optimization. J Sep Sci. 2008;31:3809-16. PMID: 19021167. DOI: 10.1002/jssc.200800455.

Read also