How Laboratories Evaluate Peptide Purity: A Technical Overview

What this peptide purity overview is for

Laboratories that buy lyophilized research peptides eventually have to defend a number. A product card says ≥99% HPLC. A lot PDF prints 99.4% at 220 nm. A PI asks whether that means the vial is 99.4% peptide by weight. A purchasing officer asks whether the file meets ICH language they saw on a biologics specification. This page answers those questions as analytical practice for Research Use Only reagents. It does not provide dosing, injection, cycles, or personal-use mixing instructions, and it does not turn a storefront COA into a licensed-drug specification.

The companion hub articles already cover two slices of this problem. How to read a research peptide COA is the receiving walkthrough: match the vial, find the fields, file the PDF. HPLC purity for research peptides is the percentage explainer: area-percent is useful, limited, and easy to over-read. This technical overview sits underneath both. It walks through the methods a competent peptide laboratory actually runs — RP-HPLC, LC-MS, amino-acid analysis, Karl Fischer, ion chromatography, and bacterial-endotoxin testing — and the impurity classes those methods are meant to see.

ICH Q6B (Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products) is the public framework many QA groups reach for when they write in-house confirmation methods. Q6B was written for licensed biologics, not for catalog RUO peptides. We cite it because the test menu it describes — identity, purity, impurities, quantity, process residuals — is the same menu a careful research lab uses when it decides what “evaluate purity” means. A research COA that only prints HPLC area-percent is a lean subset of that menu, not a claim that Q6B acceptance criteria have been met.

Nextday Peptides posts vendor and contract-lab lot files in the COA Library. We do not independently test each lot in-house. Methods and results remain those of the issuing laboratory. Your job at receiving is to read the file that belongs to the bottle you are holding, not to treat a catalog banner as a substitute for that file. Everything below assumes institutional laboratory context.

RP-HPLC C18 purity: TFA/acetonitrile and 214/220 nm area-percent

Reversed-phase HPLC on an alkyl-bonded silica column — almost always C18 for mid-length research peptides — is the workhorse purity method. The mobile phase is typically water and acetonitrile with a low percentage of trifluoroacetic acid (TFA) as an ion-pairing agent. TFA protonates residual silanols, pairs with basic residues, and sharpens peaks. Detection is UV at 214 nm (peptide-bond absorbance) or 220 nm (a slightly cleaner baseline on some systems). Integration software assigns area to each peak the method treats as relevant and reports the main analyte as a percentage of that total. That percentage is the number printed on most research-peptide COAs.

Area-percent is not a gravimetric assay. It is a relative measurement inside the detector’s response window. Species that do not absorb at the monitored wavelength, or that co-elute with the main peak, sit outside what the number “sees.” Early-eluting salts and leftover TFA often have little UV response at 214 nm. Water has none. Closely related deletion sequences can hide under the main peak if the gradient is too steep or the column is too short. D’Hondt and colleagues’ 2014 review of related impurities in peptide medicines is still the practical catalog of what SPPS actually leaves behind — and why a single C18 method is necessary but not sufficient.

Method footnotes matter more than the headline. Column chemistry (C18 versus C8 or phenyl), particle size, pore size, gradient slope, flow rate, column temperature, sample solvent, injection mass, and the laboratory’s integration rules all move the percentage. Two files that both say “99.2% HPLC” are not interchangeable evidence if one used a 5–95% acetonitrile gradient in 10 minutes and the other used a shallow 20–40% window around the main peak. When your protocol is sensitive to a known process impurity, ask whether the reported method is expected to resolve that species.

Detector saturation is the quiet way a main peak becomes artificially dominant. Overloaded injections flatten the apex and swallow shoulders. You will not always receive raw chromatogram files as a buyer, but a report that looks oddly perfect — no minor peaks labeled, no method header, no wavelength — is a reason to request a cleaner PDF rather than to invent an interpretation. Copy the printed string, including any ≥ sign, into the notebook. Do not average ≥99% threshold claims with measured 99.4% values in a trend chart.

For long acylated incretin analogs such as Retatrutide (GLP3-Reta) research peptide, lipophilicity and self-association can shift retention and peak shape relative to a 10-residue fragment. A method tuned for BPC-157 is not automatically valid for a 39-residue fatty-diacid conjugate. Compound-class methods belong on the COA. Generic “HPLC purity” without a method ID is a weaker document.

LC-MS identity: the main peak still needs a name

Purity answers “how clean under this method.” Identity answers “is the main peak the molecule we think it is.” Liquid chromatography–mass spectrometry is the orthogonal check most research-peptide files use. Electrospray ionization of a peptide typically produces a family of multiply charged ions. Deconvolution yields an observed average or monoisotopic mass. The issuing lab compares that mass to the theoretical mass of the intended sequence, including the expected salt form and any stated modifications (DAC, fatty-diacid acylation, C-terminal amide, disulfide).

A mass match within the lab’s stated tolerance supports identity. It does not by itself prove sequence order. Isobaric swaps (Leu/Ile), some deamidation events (Asn to Asp is +1 Da, often resolved), and certain deletion-plus-adduct coincidences can fool a low-resolution scan. High-resolution MS or MS/MS sequence tags close more of those gaps when the protocol is sequence-critical. Lean research COAs often stop at a single observed molecular ion. Know which level you received.

LC-MS also sees impurities the UV trace under-reports. A deletion sequence that co-elutes at 220 nm can still appear as a second mass in the extracted-ion chromatogram. Oxidation (+16 Da), TFA adducts, and incomplete deprotection (leftover t-butyl or Pbf masses) are the usual suspects we see discussed on vendor files. If the identity section lists only the main ion and is silent on related masses, you do not have an impurity map — you have a name check.

Catalog nicknames make identity errors worse. A storefront title for CJC-1295 with DAC versus CJC-1295 without DAC is a modification difference of several kilodaltons in practical terms (the DAC maleimidopropionyl-lysine adduct). HPLC purity alone will not rescue a mix-up. File the observed mass next to the intended sequence. See catalog names, CAS, and peptide identifiers when synonym collisions are the immediate problem.

If the observed mass refuses to match, stop. Do not “correct” the theoretical mass until it agrees. Request the lot file that belongs to the vial, or escalate to orders@nextdaypeptides.com with the order number and a label photo. Identity failures at receiving are cheaper than identity failures in a methods-section revision.

Amino-acid analysis and net peptide content

HPLC area-percent describes the organic-peptide peak family the UV method can see. It does not tell you what fraction of the powder’s weighable mass is peptide. Lyophilized cakes contain water, residual solvent, and a counterion — commonly trifluoroacetate after TFA-based cleavage and HPLC, sometimes acetate after salt exchange. Amino-acid analysis (AAA) after acid hydrolysis is the classical way to count residues and, with a gravimetric sample, to estimate net peptide content.

Net content is the number that converts “we weighed 2.0 mg of powder” into “we have X mg of peptide sequence in this volumetric flask.” Assay-standard arithmetic for in-vitro work belongs in your institutional method. We do not publish personal-use reconstitution calculators. If your SOP uses net content to prepare a molar standard, record the AAA or elemental result from the lot file, not a guessed 80% rule of thumb copied from a forum.

AAA has blind spots. Tryptophan is destroyed under standard 6 N HCl hydrolysis. Cystine/cysteine recoveries are method-dependent. Serine and threonine partially degrade. Asparagine and glutamine hydrolyze to Asp and Glu, so the chromatogram cannot distinguish the amide from the acid. A competent AAA report states the hydrolysis conditions and which residues were used for the calculation. A single “peptide content 85%” line without that footnote is a weaker document.

Nitrogen assay (elemental analysis) is the other common net-content route on small-molecule-adjacent files. It answers a similar weighable-mass question with different assumptions. Do not mix AAA percentages and nitrogen-assay percentages in one trend chart without noting the method. Both are still silent on whether the nitrogen belongs to the intended sequence or to a deletion cousin.

Many lean RUO COAs omit net content entirely. That is common, not automatically fraudulent. If your protocol needs it, put AAA or a stated counterion-and-water panel in the purchase specification. Specifications for research reagents are buyer-defined. A storefront HPLC banner does not imply a full Q6B quantity test.

Karl Fischer water and ion chromatography for TFA or acetate

Water is the mass you weigh that is not peptide and not salt. Karl Fischer titration remains the reference method for residual water in lyophilized solids. Volumetric KF suits higher water; coulometric KF suits low residual moisture. The number matters for two reasons: it adjusts net content, and it flags cakes that picked up humidity after lyophilization or during poorly sealed storage. A high KF result on a lot that otherwise looks clean is a receiving note, not a reason to invent a purity scandal from the HPLC line alone.

TFA is the other mass you almost always have if the peptide was cleaved and purified in TFA systems. Ion chromatography (or a calibrated 19F NMR method in some labs) quantifies trifluoroacetate. Acetate appears when the vendor performed a salt exchange. The counterion is not an “impurity” in the HPLC-purity sense; it is part of the isolated salt. It becomes a problem when a protocol assumes the free base, when TFA concentration affects a cell-based readout, or when two lots have different salt forms and nobody wrote it down.

Vergote and colleagues’ 2009 regulatory-pharmaceutical overview of peptide quality specifications is still a useful reminder that identity, purity, and quantity are separate acceptance concepts. Counterion and water live under quantity and process residuals, not under UV area-percent. Labs that treat “99% HPLC” as “99% of the bottle is peptide” are making a unit error.

Residual organic solvents (cleavage cocktails, DMF, acetonitrile) may appear on a residual-solvent panel when the vendor runs one. ICH Q3C is the licensed-drug framework for those limits. RUO files often skip the panel. If your institution requires it for a particular study, specify it. Do not infer absence of solvent from absence of a line on a one-page COA.

Record salt form in the LIMS the same way you record lot number. GHK-Cu adds a copper-stoichiometry question on top of the usual counterion story. A copper peptide COA that is silent on metal content is incomplete for any protocol that cares about Cu(II) equivalents. Ask for the metal assay when the method depends on it.

Bacterial endotoxin (LAL) and what it does not prove

Limulus amebocyte lysate (LAL) testing estimates gram-negative bacterial endotoxin in a sample, usually reported in EU/mg or EU/vial. USP <85> and the FDA’s pyrogen-and-endotoxins Q&A are the public method references most contract labs point to. Some research-peptide COAs include a LAL line because the destination lab’s cell-culture SOP has an endotoxin gate. Many lean files omit it.

Endotoxin is not HPLC purity. A chromatographically clean peptide can still carry endotoxin if water, glassware, or downstream handling introduced lipopolysaccharide. Conversely, a lot with a modest HPLC impurity can still pass a loose LAL specification. Do not use one number as a proxy for the other. If your culture work requires a stated EU limit, put that limit in the purchasing brief and confirm the COA units before you thaw cells.

LAL has interference modes. Some peptides inhibit or enhance the clotting reaction. A competent report states the method (gel-clot, turbidimetric, chromogenic), the dilution, and whether a positive-product-control spike recovered inside the acceptance window. A bare “<1 EU/mg” with no method is weaker evidence than a spike-recovered result.

Endotoxin testing is not sterility testing, not a bioburden count, and not permission to treat a research vial as an injectable drug product. HPLC plus LAL plus a pretty cake still describe a research reagent. Nextday Peptides sells Research Use Only material not for human or veterinary use. A LAL line does not change that lane.

If a file is silent on endotoxin and your SOP requires it, you have three honest options: request the test, run it in-house or at a contract lab under your quality system, or choose a lot that already carries the panel. Silent omission is not a pass.

Deletion sequences, oxidation, and deamidation on SPPS lots

Solid-phase peptide synthesis fails in characteristic ways. Incomplete coupling produces deletion sequences missing one residue. Incomplete deprotection leaves t-butyl, Trt, Pbf, or Boc masses on the product. Acid-catalyzed aspartimide and subsequent iso-aspartate formation scramble aspartic-acid sites. Methionine and tryptophan oxidize. Asparagine and glutamine deamidate, especially in solution and at elevated pH or temperature. D’Hondt’s impurity review groups these as synthesis-related versus degradation-related families. A purity evaluation that cannot name which family it is looking at is just a percentage.

Deletion sequences are the reason shallow gradients and orthogonal MS matter. A des-Ser analog of a 15-residue peptide can co-elute with the parent on a fast C18 screen and still wreck a sequence-critical binding assay. If your protocol cannot tolerate a known deletion, ask whether the vendor method resolves it, or run a confirmatory LC-MS under your own SOP. Do not assume ≥99% means every possible des-residue species is below 1%. The 1% remainder is a pool of whatever the method can see, not a certified list.

Oxidation is time- and oxygen-dependent. Lots that sat warm, or working solutions that sat on the bench, accumulate +16 Da (sulfoxide) and further products. A receiving HPLC that matches the COA does not freeze the lot in time. Storage history belongs next to the purity number when you troubleshoot a later control failure. See the lyophilized-storage article on this hub for the physical side of that history.

Deamidation is the slow leak in Asn- and Gln-rich sequences. It can look like a shoulder on the main peak or like a distinct later-eluting acid. Mass shift is +1 Da per event, which low-resolution MS can miss if the operator is not looking. High-resolution MS or a charged-variant method (ion exchange, icIEF for longer chains) catches more of it. For research catalog peptides, the practical habit is: note whether the sequence is deamidation-prone, keep solution holds short under your SOP, and treat a new shoulder on a previously clean lot as a document-and-investigate event.

Insertion sequences, racemization at activated residues, and leftover scavenger adducts (thioanisole, water, indole) complete the usual list. You do not need to hunt every species on every RUO lot. You do need to know which species your assay cares about and whether the published method can see them. That is how laboratories evaluate peptide purity in practice: a risk question, not a single banner percentage.

• Name the impurity family you care about before you argue about 0.2% on a COA
• Ask whether the HPLC method is expected to resolve that family
• Use LC-MS extracted ions when UV shoulders are ambiguous
• Record storage history next to any new oxidation or deamidation shoulder
• Do not treat the 1% remainder as a certified impurity list

Methods table laboratories use to evaluate peptide purity

The table below is the menu we walk purchasing and receiving staff through when a protocol says “confirm purity” and nobody has defined the phrase. It is a documentation aid, not a claim that every Nextday Peptides lot file contains every row. Lean research COAs commonly include HPLC plus MS. Water, counterion, AAA, and LAL appear when the vendor or the buyer specified them.

Download the lot PDF from the COA Library and tick these rows against what the file actually contains. Missing rows are a specification conversation, not a reason to invent passing results. If you later run in-house confirmation, write the method ID next to the vendor result so a future reader can see which number came from which laboratory.

How ICH Q6B language maps onto research-peptide COAs

ICH Q6B organizes specifications for biotechnological and biological products into identity, purity and impurities, potency, quantity, and process-related residuals, with method validation expected under the ICH Q2 family. Licensed peptide drugs and recombinant proteins live in that world. Catalog RUO peptides do not. The useful move for a research lab is to borrow the category names without borrowing the legal status.

Map the COA onto those categories in the notebook. Identity: MS and/or retention-time match. Purity: RP-HPLC area-percent. Impurities: whatever related peaks or masses are listed (often none on a lean file). Quantity: fill weight plus, when present, net content. Residuals: water, TFA, solvents, endotoxin. Potency, in the Q6B sense of a functional bioassay, is usually absent from research-catalog files and should not be inferred from HPLC.

Q6B also distinguishes product-related impurities (variants of the intended sequence) from process-related impurities (host-cell protein, DNA, reagents). For SPPS research peptides the “host cell” is a resin and a set of chemicals. Process residuals are leftover scavengers, metals, TFA, and water. Product-related impurities are deletions, oxidations, and deamidations. Saying that out loud in a lab meeting prevents people from treating every extra HPLC peak as “dirt from the factory.”

Method validation is the other Q6B/Q2 expectation that RUO files rarely meet in full. Specificity, linearity, accuracy, precision, range, and robustness are what a regulated lab documents for a release method. A contract-lab research COA may state a method ID without a validation summary. If your quality system requires validated confirmation, that work happens at your institution or at a lab you qualify. Storefront education cannot certify a method for every destination lab.

Use the Q6B PDF as a vocabulary source when you write an internal specification. Do not tell an auditor that a research lot is “Q6B compliant” because the HPLC number looks high. That sentence is almost always false and always unhelpful.

Receiving habits that make purity data usable

Purity evaluation fails as often from paperwork as from chemistry. A perfect chromatogram attached to the wrong lot string is worse than a lean file attached to the right vial. Start with the match: product name, fill amount, lot number on the label versus the PDF header. The COA reading guide is the step-by-step; this paragraph is the reason it exists.

Name the saved PDF with the lot string. Copy the HPLC value exactly. Note wavelength and method ID when printed. File identity mass next to theoretical mass. Tick water, counterion, AAA, and LAL as present or not specified. Assign a storage location. That checklist takes minutes and saves weeks when a control drifts.

Blend SKUs need an extra question: does the file cover the blend or each component? A single HPLC percentage on a multi-sequence vial is a different claim than three component files. Ask before the blend enters a study freezer. Ambiguity here is how labs invent a combined purity number that no chromatogram ever measured.

Nextday Peptides fulfills from U.S. warehouses in Florida, North Carolina, and California with a same-day processing target for orders placed before 3:30 PM Eastern, Monday–Saturday, after payment clears. Transit is not a substitute for your receiving SOP. Move vials into labeled storage and file the COA promptly. Details live in the shipping and delivery policy.

When staff leave, export the lot ledger. Institutional knowledge that “we always trust the 99% banner” is not a purity system. Cross-link purchase-order numbers to lot files so auditors see a closed loop: ordered, received, filed, stored, used. Browse the products catalog with the same filter: intended use first, then identity, then the method menu you actually need.

• Match vial lot to PDF header before you open the chromatogram discussion
• Record method-bound numbers with their method IDs
• Treat missing panels as unspecified, not as passing
• Separate blend paperwork questions from single-sequence lots
• Keep RUO labeling intact on any lab-owned secondary container

When labs run their own confirmation methods

Some institutions accept vendor COAs as receiving evidence. Others require an in-house or third-party confirmation before a lot enters a regulated study folder. Both postures are legitimate. What fails is a silent hybrid: the lab says it “checks purity” but has no written method, no system suitability, and no record of what it would do if the result disagreed with the vendor file.

A minimal confirmation method for a research peptide is usually a C18 screen plus a mass check, with system suitability on a retained reference or on a well-characterized prior lot. Write the gradient, the wavelength, the integration rules, and the mass tolerance. Run a blank and a suitability injection. Decide in advance what constitutes a mismatch worth quarantining. Hope is not a specification.

Disagreements have ordinary causes: different columns, different integration of a shoulder, sample-prep loss, overloaded vendor injections, or an actual different lot. Investigate before you assume fraud or assume your method is wrong. Photograph the vial, re-download the COA Library file, and compare method IDs side by side. Escalate with evidence, not with a Slack screenshot of one percentage.

In-house confirmation is still RUO documentation. It does not convert the material into a drug substance and it does not authorize administration to people or animals. Keep the use restriction in the same notebook entry as the chromatogram. Nextday Peptides sells Research Use Only material; your confirmation method does not change that.

Purity claims that look careful but are not

Treating ≥99% HPLC as net peptide content is the most common unit error we see in receiving tickets. Area-percent is not weighable-mass percent. Water and TFA still sit in the cake.

Treating a clean MS ion as proof that deletion sequences are absent is the second. Identity of the main peak is not a full impurity profile.

Treating LAL as sterility, or as permission to inject, is a category error with compliance consequences. Endotoxin is one residual. RUO is the use lane. They do not trade.

Treating two lots as interchangeable because both banners say 99% is how assays drift. Methods, salt forms, and storage histories differ. Compare files, not adjectives.

Treating social-media “purity tests” filmed on kitchen counters as laboratory evaluation is outside the scope of this catalog. If the work is not institutional research, do not buy these materials.

Bottom line for laboratories evaluating peptide purity

Evaluate peptide purity as a menu, not as a banner. RP-HPLC C18 area-percent at 214 or 220 nm is the headline. LC-MS names the peak. AAA and counterion/water panels explain the weighable mass. LAL answers an endotoxin question when your SOP asks it. Deletion, oxidation, and deamidation are the impurity families worth naming before you argue about a tenth of a percent.

Read the hub COA and HPLC articles for field-level and percentage-level detail. Use this page when someone asks which methods exist and what each one cannot see. File every lot from the COA Library. Keep the Research Use Only boundary in the same sentence as the purity number.

Nextday Peptides supplies RUO materials and vendor lot documentation for that laboratory lane. Purity literacy after delivery is shared work: we post the files; your lab owns the specification, the confirmation method, and the notebook. If you take only one operational upgrade from this page, make it a written method-menu habit that survives staff turnover.