The Mass Spec Tolerance Red Flag Most COA Guides Miss

One line on the certificate decides whether identity was confirmed

Every guide to reading a peptide certificate of analysis tells you to check that the observed molecular weight matches the theoretical one. Almost none of them tells you how closely it has to match, and that omission is where a real defect hides in plain sight.

The tolerance line is the check. It takes about ten seconds, it requires no laboratory background, and it separates a certificate that confirms identity from one that merely appears to.

What the mass spectrometry section is establishing

Purity and identity are different questions answered by different instruments.

Chromatography answers how much of the sample is the target compound. Mass spectrometry answers whether the target compound is the one you ordered, by measuring molecular weight and comparing it against the theoretical value calculated from the amino acid sequence.

A certificate should therefore show three things in this section: the theoretical molecular weight, the observed molecular weight, and the tolerance within which those two are considered to agree.

Most buyers check that the first two numbers look similar and move on. The third number is the one that determines whether "similar" means anything.

Daltons and percentages are not interchangeable

Here is the specific failure.

Molecular weight is expressed in daltons. A tolerance can be stated in daltons, or it can be stated as a percentage of the molecule's mass. Those two ways of writing a tolerance sound equivalent and are not, because a percentage scales with the size of the molecule.

Take a peptide of roughly 3,000 daltons, which is an ordinary size in this category.

A tolerance of plus or minus 0.5 daltons permits a variance of half a dalton. That is a realistic specification for a high-resolution instrument, and on a standard instrument plus or minus 1 dalton is a defensible one.

A tolerance of plus or minus 0.1 percent, on that same 3,000 dalton peptide, permits a variance of about 3 daltons.

Three daltons is a different order of claim. It is six times wider than the half-dalton specification and three times wider than the standard-instrument one, and it is wide enough to accommodate defects the test exists to detect.

What fits inside three daltons

What fits inside a three dalton window is the part that matters, because a wide tolerance is only a problem if something real can hide in it.

Amino acid residues differ from each other by small mass differences. Some pairs differ by a single dalton or less. A sequence that has substituted one residue for a similar one, or that has lost or gained a small modification during synthesis, can produce an observed mass that sits inside a three dalton window while being the wrong molecule.

Truncation is the more consequential case. A synthesis that failed to add a residue produces a shorter peptide. Depending on which residue is missing, that mass difference can be large and obvious, or it can be small enough that a generous tolerance reports agreement.

The certificate then says, correctly by its own stated criterion, that identity was confirmed. The criterion was simply loose enough that confirmation did not require the molecule to be right.

Why instrument capability is the benchmark

The reason a percentage tolerance reads as a red flag rather than merely a stylistic choice is that it is far wider than what the instruments actually achieve.

Modern liquid chromatography mass spectrometry systems deliver mass accuracy in the range of parts per million. Expressed against a 3,000 dalton peptide, single digit parts per million corresponds to hundredths of a dalton.

A laboratory specifying a tolerance of three daltons is therefore specifying a window roughly two orders of magnitude wider than its own instrument's capability. There is no analytical reason to do that. The instrument is not the constraint.

Which leaves two explanations, and neither is reassuring for a buyer. Either the specification was copied from a template without anyone considering what it permits, or it was set wide enough that borderline material passes.

How to run the check

The procedure is short.

• Find the tolerance line. It sits with the mass spectrometry result, often phrased as agreement within, or as a specification.

• Note the unit. Daltons or a percentage. This is the whole check.

• If it is a percentage, convert it. Multiply the theoretical molecular weight by the percentage. On a 3,000 dalton peptide, 0.1 percent is 3 daltons.

• Compare against the residue masses that matter. If the permitted window is larger than the mass difference of a plausible substitution or truncation, the test cannot exclude that defect.

• If no tolerance is stated at all, treat identity as unconfirmed. A comparison with no acceptance criterion is not a test, it is two numbers printed near each other.

What to ask the supplier

A supplier who understands their own documentation can answer these without hesitation. One who cannot is telling you something about how the certificate was produced.

• What tolerance does your laboratory apply for mass agreement, and in what unit?

• What instrument class performed the analysis?

• Can you provide the raw spectrum rather than only the reported value?

• Who set the acceptance criterion, your laboratory or the manufacturer?

The last question is the interesting one. An acceptance criterion set by the party being tested is not an independent criterion, which is the same structural problem that runs through what third-party tested actually means.

The wider pattern this belongs to

The tolerance issue is one instance of a general pattern in analytical documentation, and recognising the pattern is more valuable than memorising the instance.

Every measurement on a certificate has an acceptance criterion, stated or unstated. The number is the result. The criterion is what makes the result mean something. A document that reports results generously and criteria vaguely can be entirely accurate in every number it prints and still fail to establish what it appears to establish.

Purity has the same structure. A percentage with no named chromatographic method is a result with no criterion. The reader sees 99 percent and supplies the missing rigour from imagination.

This is why reading a certificate diagnostically rather than descriptively matters, and why the full reading procedure is worth having: how to read a peptide certificate of analysis.

What this check does not tell you

Being precise about the boundary matters as much as the check itself.

A tight mass tolerance confirms that the molecule measured is very likely the one specified. It says nothing about how much of the sample that molecule represents, which is the purity question. It says nothing about sterility, endotoxin, residual solvents or handling after testing. And it says nothing about what any compound does in a person, which is not a question analytical chemistry answers.

A certificate with an excellent mass tolerance and no purity method is still an incomplete document.

Why identity confirmation carries more weight than it looks like it does

The tolerance question can seem like pedantry until you see how seriously identity and related-substance characterisation are handled where the stakes are formal.

FDA guidance on abbreviated applications for certain highly purified synthetic peptide drug products expects an applicant to show that the product does not contain any new specified peptide-related impurity above 0.5 percent of the drug substance, to characterise each such impurity, and to justify why those at or below that level are acceptable. The guidance is published at FDA guidance on ANDAs for certain highly purified synthetic peptides.

Read that against a wide mass tolerance and the tension is obvious. A framework that asks for each related substance to be identified and justified is incompatible with an identity check loose enough that a substituted residue passes as agreement. The two are answering the same underlying question, which is what exactly is in this material, and one of them answers it and the other does not.

Again, this framework governs drug approval and not research supply. The point is calibration. When the consequences are formal, a fraction of a percent of a structurally related molecule is something a framework asks you to name. A certificate that cannot exclude a whole substituted residue is a long way from that standard.

What the tolerance means for the seller

A tolerance is not only a buyer's concern. For a business publishing a purity or identity claim, the specification behind that claim is the evidence for it.

The FTC's Health Products Compliance Guidance requires substantiation in the form of competent and reliable scientific evidence, held at the time a claim is made. A certificate whose acceptance criterion is wider than the defect it is meant to exclude is weak substantiation for a claim about identity, and it is weak in a way that is visible on the face of the document to anyone who checks the unit.

Frequently asked questions

Is a percentage tolerance always wrong?

Not automatically wrong, but it is nearly always wider than instrument capability on a peptide-sized molecule, and it should prompt the question of why it was set that way. Convert it to daltons and judge the resulting window.

What tolerance should I expect to see?

Plus or minus 0.5 daltons is a realistic specification for a high-resolution instrument, and plus or minus 1 dalton for a standard one. Both are defensible. Both are far tighter than a percentage specification on a molecule of this size.

The certificate has no tolerance line at all. What now?

Ask for it. Until it exists, the identity comparison has no acceptance criterion and the confirmation cannot be evaluated.

Does this apply to small peptides too?

The arithmetic changes with molecular weight, which is the point. A percentage tolerance permits a narrower absolute window on a smaller molecule and a wider one on a larger molecule. Always convert to daltons before judging it.

Is this article legal or medical advice?

No. It describes how to evaluate an analytical specification. Compounds in this category are supplied for laboratory research purposes only.

Where this fits

The full certificate reading procedure is in how to read a peptide certificate of analysis, and the operational version, as questions to ask before a first order, is in the supplier vetting checklist.

All compounds referenced anywhere on this site are supplied strictly for laboratory research purposes only. Nothing here is for human consumption, and nothing here is intended to diagnose, treat, cure, or prevent any disease.