qNMR vs mass balance: how purity is really assigned
A purity figure is a measurement result, not a property of the material, and the route used to obtain it decides what the number can be used for. This compares the two assignment routes that dominate reference standards - mass balance and quantitative NMR - and follows either one through to a weighing.
One compound, two purity values
Two suppliers send certificates for the same CAS number. One states 99.4 percent, the other 97.8 percent. Both documents can be honestly produced, internally consistent and defensible. The gap is usually not competence; it is the measurement model behind the number.
A purity value answers a narrower question than most buyers assume. It is the output of a defined procedure applied to a defined measurand, on a defined basis: change any of the three and the number moves while the material does not.
What a mass balance adds up
The mass balance route works by subtraction. The main component is taken to be whatever is left once everything else has been measured:
mass fraction of main component = 100 percent - the sum of the mass fractions of all other components
| Term in the balance | Typical technique |
|---|---|
| Organic impurities, including related substances and degradants | HPLC with UV detection, or GC with flame ionisation detection |
| Water | Karl Fischer titration |
| Volatile content, including residual solvents | Thermogravimetric analysis, loss on drying, or headspace GC |
| Inorganic residue and counter-ion | Residue on ignition, sulphated ash, titration or ion chromatography |
The subtraction shape has a consequence that is easy to miss: every term the balance does not look for is added to the main component. A balance that omits water, or the counter-ion of a salt, is not a smaller measurement of the same thing. It is a biased one, and the bias always runs in the same direction - towards a flattering number.
Two assumptions hold the arithmetic together: that every impurity present has been detected and quantified, and that each detected species produces a response proportional to its mass. The second is the weaker, and it is where most disagreement between suppliers originates.
Why area percent is not mass fraction
A chromatographic purity of 99.5 area percent is a statement about detector response, not about mass. An impurity with a weaker chromophore than the main component is undercounted; one with a stronger chromophore is overcounted. Converting area percent into a mass figure requires a relative response factor for each impurity, and measuring that factor requires the impurity itself as a characterised material - which is one reason response factors are frequently taken as unity instead.
Three categories of material are effectively invisible to a UV chromatogram: water, residual solvents, and inorganic residue including the counter-ion of a salt. A lot that is 98.0 percent main component by mass, with 1.5 percent water and 0.5 percent residual solvent, can still print 99.9 area percent, because neither water nor solvent absorbs at the detection wavelength. The chromatogram is not wrong. It is answering a different question from the one the buyer is asking.
What qNMR measures instead
Quantitative proton NMR with an internal standard is a direct ratio measurement. The signal intensity of a chosen proton environment in the analyte is compared with the signal from a known mass of an internal standard whose own purity is known. Signal intensity is proportional to the number of contributing protons, and the natural abundance of proton is effectively 100 percent, so the proportionality does not depend on a compound-specific calibration.
Three things follow from that. The result is traceable to the SI through the mass measurements and the internal standard; the main component is measured directly rather than inferred by subtraction; and the same spectrum carries identity information, so the assignment and the structural confirmation come from one experiment.
The demands are equally specific. The preparation has to be gravimetric, because every mass error enters the result directly. The internal standard has to be soluble in the same solvent and of known purity - an impurity in it becomes an error in the analyte. Signals for the chosen protons have to be resolved and integrable, which rules out overlapping structural relatives, and exchangeable protons need care because their signals move with water content. As with any quantitative procedure, the method is expected to be validated for its intended use.
Why the qNMR figure is often the lower one
NIST compared the two routes on 25-hydroxyvitamin D reference standards and reported a consistent pattern: qNMR results were often lower than the mass balance estimate. The reading is not that NMR measures low. It is that an incomplete balance overstates the main component, because whatever the balance failed to see stays in the total.
That study needed several techniques together - qNMR, liquid chromatography with UV and with mass spectrometric detection, Karl Fischer titration and thermogravimetric analysis - and found volatile and water impurities alongside structurally related impurities that chromatography struggled to observe and that one-dimensional NMR could not distinguish from the main species. The conclusion was not that one method wins, but that combining a direct measurement with an informed balance is what removes the bias.
For a buyer this changes the question. The two numbers are not competing claims about one quantity; they are the outputs of two different models. A supplier who can say which model produced the figure, and why it suits the intended use, is more useful than one quoting a higher number.
What a well-stated value contains
A certified value from a national metrology institute shows the full form. NIST SRM 921a, cortisol, states a certified mass fraction of 99.3 percent with an expanded uncertainty of 0.4 percent at 95 percent confidence, and a standard uncertainty of 0.2 percent. The measurand is named as the mass fraction of cortisol, expressed as a percentage. Traceability is stated to the SI through the realisation of the unit for chemical mass fraction. The value was assigned by quantitative proton NMR as a primary ratio measurement. The certificate carries an expiry date, storage instructions, and the statement that certification is nullified if the material is damaged, contaminated or otherwise modified.
Four elements in that paragraph are worth checking on any certificate: what is being measured and on what basis, how it was assigned, an uncertainty with a stated coverage, and a validity limit tied to handling. Without them a figure is a number rather than a specification.
A pharmacopoeial reference standard makes a different trade, deliberately. The Ph. Eur. information leaflet states that the unit quantity printed on the vial is the approximate amount of material filled into it and is not to be considered accurate from an analytical point of view, and it may declare that no uncertainty is stated because it is negligible against the defined limits of the method-specific assays the standard is used for. Validity is handled through a batch validity statement checked at the time of use. Both models are fit for purpose, but they are not interchangeable.
The basis decides the arithmetic
Purity without a basis is ambiguous, because the same lot can be described on several:
| As-is | The material as supplied, water and residual solvent included in the figure. The number that matches what sits on the balance. |
|---|---|
| Dried or anhydrous | Corrected for water and volatiles. Higher than the as-is figure, and not directly usable for weighing unless the water content is applied separately. |
| Free base equivalent | Corrected for the salt or hydrate. The right basis for comparing two suppliers who ship different salt forms of the same API. |
| Vial fill | The approximate quantity dispensed into the container. An inventory figure, not an analytical one. |
A worked example, with a hydrochloride salt standard whose as-is content is 98.0 percent, a free base molecular weight of 300.0 and a salt molecular weight of 336.5. The target is 10.0 mg of free base equivalent in 10.0 mL of solution:
- Mass of salt that carries 10.0 mg of free base: 10.0 x 336.5 / 300.0 = 11.2 mg
- Mass to weigh, correcting for the assigned content: 11.2 / 0.980 = 11.5 mg
Weigh 11.5 mg and make up to volume. Had the certificate stated the content on a dried basis, the water content would have to be applied separately. And if the material takes up moisture while it is being weighed, the balance no longer agrees with the certificate - which is why hygroscopic standards are weighed quickly, in a controlled atmosphere, or handled as solutions.
What to ask before you buy
- What exactly does the assigned value measure, and on what basis - as-is, dried, free base, or the salt form?
- Which procedure assigned it, and was a second, independent procedure used to check it?
- Is water accounted for, by what method? Are residual solvents and inorganic residue accounted for?
- Is an uncertainty stated, with its coverage and confidence level? If none is stated, on what basis is it considered negligible for your intended use?
- Through what is the value traceable?
- Is the value lot-specific, and does the lot on the certificate match the lot on the vial?
- What are the retest date and validity rules, and the storage and shipping conditions? Does the assignment survive repackaging or exposure to air?
- For a neat material used to calibrate, is the producer assessed against ISO 17034 for reference material production, and by which body?
- Will you be notified if the assigned value is revised?
Traps worth naming
- Feeding area percent into a weighing calculation. It overstates content by whatever water, solvent and response-factor error the chromatogram did not see - see what an impurity reference standard is for why the basis decides what a value can be used for.
- Reading the vial fill as an analytical quantity. It is an inventory figure, and certificates say so.
- Comparing purity across suppliers without comparing methods. A mass balance figure and a qNMR figure are not the same measurement, and the bases may differ too.
- Drying a standard whose value is stated as-is. Once the water is gone, the number on the certificate no longer describes what is on the balance.
- Assuming the higher number is the better material. On the same lot, a more complete balance usually produces the lower figure.
- Treating the certificate as a property of the catalogue number. It describes a lot - see how to read a certificate of analysis for the fields that carry from lot to lot and the ones that do not.
Where to start
What decides whether a standard can carry a quantitative method is not the headline purity but the assignment behind it: the measurand and its basis, the route used, the uncertainty, and the validity statement. Identification, a limit test, and a quantitative assay are three different requirements, and only the last one needs the full assignment package - so it is worth saying which one you need. Send the structure or SMILES, the quantity and the intended use to technical@rssechem.com. The impurity standards index is organised by parent drug substance, and the certificate of analysis page shows the fields we report.
Frequently asked
Why is qNMR purity often lower than the HPLC purity of the same material?
Can I use the area percent from a chromatogram as the content of my standard?
What does a statement such as 99.3 percent plus or minus 0.4 percent mean?
Should I dry a reference standard before weighing it?
Which is better, qNMR or mass balance?
Does the quantity printed on the vial tell me how much material I have?
Sources and further reading
Every document below is published by the body that owns it; none of them is a summary written by us. Accessed 2026-10-15.
- NIST — Quantitative NMR and mass balance purity assessment of 25-hydroxyvitamin D reference standards (NIST publication 918319) — the mass balance definition and its two assumptions, the statement that congruent response factors are often assumed erroneously, the finding that qNMR values are often lower, and the orthogonal method set of qNMR, LC-UV, LC-MS, Karl Fischer and TGA
- NIST — Certificate of Analysis, SRM 921a Cortisol — the worked example of a certified value: a certified mass fraction with expanded and standard uncertainty, the named measurand, SI traceability, assignment by qNMR as a primary ratio measurement, an expiry date and the handling conditions that nullify certification
- NIST — Methods for the SI value assignment of the purity of organic compounds (IUPAC technical report) — the metrological framework for assigning the purity of an organic material directly, rather than by subtracting everything else
- ICH — Q2(R2) Validation of Analytical Procedures — the validation expectations that any quantitative procedure has to meet, including the procedures used to assign or verify a purity value
- EDQM — Information Leaflet, Ph. Eur. Reference Standard — the statements that the unit quantity on the vial is approximate and not analytically accurate, that an uncertainty may be omitted because it is negligible against the limits of the intended assay, and that validity is checked through a batch validity statement
- ISO — ISO 17034, General requirements for the competence of reference material producers — the requirements a reference material producer is expected to meet, which is the frame a buyer can use to judge an assigned value and its documentation
Send us a CAS number
Availability, documentation and lead time — answered by the people who handle the batch, not a ticket queue.