A purity percentage measures how much of a sample is the target peptide. It says nothing about the organic solvents left behind when that peptide was synthesised and purified. Residual solvents testing is the separate assay that quantifies those leftovers, and for a material destined to be dissolved and injected it addresses a class of contaminant a purity figure structurally cannot see.
Where residual solvents come from
Synthetic peptides pass through several solvent-heavy steps before they reach a vial. Solid-phase synthesis, cleavage from the resin, and chromatographic purification all use organic solvents, and some of that solvent stays trapped in the final lyophilised powder. Reversed-phase HPLC purification of synthetic peptides typically runs mobile phases of 0.1% trifluoroacetic acid in water and 0.1% TFA in acetonitrile, where the TFA acts as an ion-pairing reagent that improves peptide peak shape (post-cleavage purification and analysis). Acetonitrile, methanol, dichloromethane, and other solvents appear across the workflow. Residual solvents testing measures how much of each remains.
This is a distinct question from purity. A chromatographic purity number reports the fraction of the peptide-related material that is the target rather than a peptide impurity, but residual solvents are small volatile molecules that a standard purity method may not register at all. Understanding what a purity percentage does and does not include is the first step to seeing why a second, solvent-specific assay exists.
The ICH Q3C framework: three classes of solvent
The governing reference is ICH Q3C(R8), the Guideline for Residual Solvents, which sorts solvents into three risk classes.
Class 1: solvents to be avoided
Class 1 solvents are known or strongly suspected human carcinogens and environmental hazards. Under ICH Q3C(R8) they should not be employed in manufacture, but if their use is unavoidable to produce a drug product with a significant therapeutic advance, their levels should be restricted to the Table 1 concentration limits unless otherwise justified. Those limits are strict: benzene 2 ppm (a carcinogen), carbon tetrachloride 4 ppm, 1,2-dichloroethane 5 ppm, 1,1-dichloroethene 8 ppm, and 1,1,1-trichloroethane 1500 ppm (included as an environmental hazard rather than for direct toxicity).
Class 2: solvents to be limited
Class 2 solvents are non-genotoxic animal carcinogens or possible causative agents of irreversible toxicity such as neurotoxicity or teratogenicity. They are controlled by a Permitted Daily Exposure (PDE) rather than banned. The relevant one for peptide work is acetonitrile, the reversed-phase HPLC mobile-phase solvent, which carries a PDE of 4.1 mg/day and an Option 1 concentration limit of 410 ppm. Other Class 2 limits in Table 2 include methanol at a PDE of 30.0 mg/day (3000 ppm), dichloromethane at 6.0 mg/day (600 ppm), hexane at 2.9 mg/day (290 ppm), and toluene at 8.9 mg/day (890 ppm).
Class 3: low toxic potential
Class 3 solvents have low toxic potential, and no health-based exposure limit is needed for them. ICH Q3C(R8) assigns them PDEs of 50 mg or more per day, and states that Class 3 residual solvents at amounts of 50 mg per day or less (corresponding to 5000 ppm, or 0.5%, under Option 1) are considered acceptable without further justification.
From a PDE to a concentration limit
A PDE is a mass per day, but a certificate reports a concentration in the powder, so the two are linked by a calculation. ICH Q3C(R8) Option 1 uses the equation Concentration (ppm) = 1000 times PDE divided by dose, with the PDE in mg/day and the dose in g/day, assuming a conservative daily product mass of 10 g. Option 2 instead applies the PDE against the known maximum daily dose of the actual product. Option 1 is the default that assumes a large 10 g intake; Option 2 lets a manufacturer with a smaller real dose justify a proportionally higher permitted concentration. Knowing which option a limit comes from is part of reading a residual-solvent result correctly.
TFA: the peptide-specific complication
Trifluoroacetic acid sits in an unusual position. It is used both to cleave peptides from the synthesis resin and as the HPLC ion-pairing modifier, yet ICH Q3C(R8) does not assign it a Class 1, 2, or 3 PDE. Instead TFA appears in Table 4, "Solvents for which No Adequate Toxicological Data was Found," a list for which manufacturers are expected to supply their own justification for residual levels. There is no tidy published ppm number to check TFA against, which shifts the burden onto the producer to characterise and defend it.
TFA is also not only a solvent here. Because of the acidic conditions of solid-phase cleavage and reversed-phase purification, synthetic cationic peptides are mainly obtained as trifluoroacetate salts, so the trifluoroacetate counter-ion has to be removed or exchanged for a more benign acid such as acetate or hydrochloride before preclinical or clinical use (counter-ion exchange evaluation, J Pept Sci). TFA is therefore both a residual process solvent and a bound counter-ion, and neither role comes with an ICH concentration ceiling to point at.
The method: static headspace gas chromatography
Residual solvents are measured by separating the volatile solvent from the non-volatile peptide, and headspace sampling does exactly that. USP General Chapter <467> determines Class 1 and Class 2 solvents by static headspace sample introduction into a gas chromatograph equipped with a flame-ionization detector (FID). The sample is warmed in a sealed vial so the volatile solvents partition into the gas above the liquid, and that headspace is injected onto the column while the peptide stays behind.
USP <467> runs three procedures: Procedure A for identification and screening on a G43-phase column, Procedure B to confirm a peak's identity on a second, orthogonal G16-phase column, and Procedure C for quantification, with the Class 2 solvents split across a Class 2 Mixture A and a Class 2 Mixture B standard solution. System suitability is defined tightly: the resolution between acetonitrile and methylene chloride (dichloromethane) must be not less than 1.0 in the Class 2 Mixture A Standard Solution, and the signal-to-noise ratio must be not less than 5 for 1,1,1-trichloroethane in the Class 1 Standard Solution. Where an orthogonal confirmation of a solvent's identity is needed, the gas chromatograph can be coupled to mass-spectrometric detection instead of, or alongside, the FID.
Why it matters for injectables
Route of administration is the reason this test exists as its own line item. A residual solvent level that would be a minor concern in an oral product is a different proposition in a material that is reconstituted and injected, where the solvent enters directly. Yet a routine analysis panel of HPLC purity plus mass-spectrometric identity does not cover residual solvents at all, which is precisely why guidance on whether research peptides are safe from a testing perspective and on how to choose a peptide testing lab both name residual solvents as one of the gaps a purity certificate leaves open. A clean purity number and a residual-solvent result answer separate questions, and only the second one speaks to what was left behind by the process.
The bottom line
Residual solvents testing is the assay that accounts for the acetonitrile, methanol, dichloromethane, and TFA used to make and purify a peptide, measured against the ICH Q3C class system by static headspace gas chromatography. Purity tells you what fraction of the material is the intended molecule; residual solvents tell you what process chemistry is still riding along with it. For an injectable context the two are not interchangeable, and TFA's place outside the ordinary class limits means the burden of proof sits with whoever made the peptide. The maintained regional adoption of this framework, EMA's ICH Q3C(R9) scientific guideline, and the broader regulatory context for therapeutic-goods quality covered at the Coalition for Better Health, both underline that a residual-solvent result is a standard expectation, not an optional extra.