Abstract - "99% of purity" is a short phrase with enormous weight, but it's often misinterpreted. In practice, this number depends on as the chromatogram was acquired (method), as has been integrated (software + parameters) e what was left out of the report (raw dataconditions, spectrum, chain of custody). This guide shows you how to evaluate baseline and integration, and lists the 6 red flags most common in "too pretty" HPLC reports. Where applicable, I also explain where LC-MS and where it doesn't work.
1) What "99%" means - 2) Baseline - 3) Integration - 4) 6 Red Flags - 5) LC-MS: where it comes in - 6) Checklist S157 - 7) Internal connections - References
If you are validating a COA for any compound in the Peptide Database, use this article as a checklist and then go through the COA Auditor for structured auditing. The S157 rule is simple: no method + no legible chromatogram + no traceability"99%" is not a fact - it's marketing.
1) What "99%" really means (and what it doesn't)
In most COAs, "99%" appears as Area% of the main peak in HPLC. This is literally the fraction of the integrated area assigned to the analyte peak, divided by the sum of the areas of all the integrated peaks in the chromatogram (under integrator rules).
- Check: that, in that method (column, gradient, solvents, detector, wavelength, run time), the dominant peak occupies ~99% of the integrated area.
- No confirmation: molecular identity (a "clean" peak can be another molecule), impurities invisible to the detector used, absence of endotoxins, and absence of manipulation via baseline/integration.
2) Baseline: the "bottom line" that decides your Area%
Baseline is the reference line of the signal when there is no relevant elution. It is never "perfect" in the real world: there is thermal noise, pressure variations, composition changes during gradients and detector limits.
2.1 Baseline "good" vs baseline "suspect"
- Good: low but existing noise, smooth transitions during gradient, no artificial "jumps", scale coherent with the signal.
- Suspect: too smooth a line (aggressive smoothing), compressed scale, baseline "stuck" to zero from start to finish, or sudden changes without explanation of method.
3) Integration: where the software becomes the judge of the result
Integration converts a trace (signal vs time) into numerical areas. The problem: there are many parameters (threshold, slope sensitivity, peak width, valley-to-valley, tangent skim, etc.). Two operators with the same chromatogram can produce Area% different.
3.1 Three questions the AOC should answer (but rarely does)
- What integration parameters were used? (or, at the very least, whether it was automatic vs. manual integration)
- Does the chromatogram show clearly integrated smaller peaks? (or were ignored/cut)
- Raw data export (CSV/ASC) or software report with visible integrations?
4) The 6 most common red flags in "99% purity"
| Red flag | What could be happening | How to validate (S157) |
|---|---|---|
| #1 - Chromatogram cut (incomplete window) | Late impurities (hydrophobic) may be outside the frame; "short run" hides peaks. | Require full race + method. Audit in COA Auditor. |
| #2 - Baseline too smooth ("zero" noise) | Aggressive smoothing, manipulated scales or "embellished" exports. | Compare scale/noise with method standards; see Lexicon: Baseline Noise. |
| #3 - Invisible integration (no marks/limits) | Smaller peaks ignored/cut off; method doesn't separate or integration was too "clean". | Request report with visible integrations; cross-reference with COA Auditor. |
| #4 - Retention Time without context | Without a column/gradient/flow, RT becomes decorative; RT "dancing" between COAs suggests inconsistency. | Require complete method; see Lexicon: Retention Time. |
| #5 - "99%" without identity | HPLC doesn't prove "what it is"; a clean peak could be another molecule. | When the risk is justified, ask for ID via LC-MS (without confusing identity with purity). |
| #6 - Lack of traceability (chain of custody) | Cute PDF not attached to the bottle/batch; may be from another sample. | Validate batch/batch + chain of custody in COA Auditor. |
5) Where LC-MS comes in: "identity" vs "purity"
LC-MS is excellent for confirming identity (mass, expected patterns, fragmentation when available). But LC-MS doesn't "solve" everything:
- MS can fail in separating isomers without adequate chromatography.
- MS may not quantify well impurities without calibration/standards.
- MS does not test for endotoxins and does not replace microbiological controls.
Reading S157: Well-done HPLC for purity + LC-MS for identity is the duo that reduces the risk of "nice peak, wrong compound".
6) Quick checklist (S157) to validate "99%"
- There is readable chromatogram (scale, real noise, full run)?
- The report includes method (column, mobile phase, gradient, flow, detector/wavelength, temperature)?
- There are smaller peaks visible and integrated (without excessive "cleanliness")?
- The lot/batch on the bottle match the batch on the COA?
- There is LC-MS (or equivalent) for identity when the risk justifies it?
- There are signs of chain of custody (traceability, independent lab, ISO/IEC 17025 when applicable)?
7) Recommended internal connections (to strengthen the web)
- COA Auditor - structured auditing of reports (method, peaks, integration, custody).
- S157 Lexicon - terms: HPLC, LC-MS, Baseline Noise, Retention Time, Area%, COA, ISO/IEC 17025.
- Lab Tools - operational context (consistency, maths and checks that avoid false conclusions).
- Peptide Database - cross compound ↔ class ↔ risk ↔ validation.
- Research Journal - S157 framework (evidence vs plausibility; confusers; documentation).
- Examples of worksheets (to practise reading COAs by class): BPC-157, TB-500, Semaglutide, Tirzepatide.
References
- Snyder LR, Kirkland JJ, Dolan JW. Introduction to Modern Liquid Chromatography. (Fundamentals of HPLC, method and interpretation.)
- Dong MW. Modern HPLC for Practicing Scientists. (Integration, parameters and pitfalls.)
- International Organisation for Standardisation. ISO/IEC 17025 - General requirements for the competence of testing and calibration laboratories(Traceability and laboratory expertise.)
- ICH Q2 (R2) / Q14 and regulatory guidelines related to validation and development of analytical methods. (General validation criteria.)
- Gross JH. Mass Spectrometry: A Textbook. Springer. (MS/LC-MS context for identity.)
Safety Note (S157): Educational content. Does not constitute medical advice. For guidance and responsible use of information, please consult the Information Use Policy.
EN 
PT 
FR 
ES 
DE 
IT 
PL