One amino acid at a time, on a plastic bead, forty chances to go wrong. The manufacturing process explains the purity ceiling, the price ladder, and most of the differences between two vials of "the same" peptide.
Before 1963, making a peptide meant solution chemistry: form one bond, then isolate and purify the intermediate — losing material at every step — then form the next. Losses compounded so brutally that peptides beyond a dozen or so residues were heroic undertakings.
Bruce Merrifield's fix, published in 1963 in the Journal of the American Chemical Society (vol. 85, pp. 2149–2154; doi:10.1021/ja00897a025 — the paper predates PubMed indexing and is cited by DOI throughout the literature), was to anchor the growing chain to an insoluble plastic bead. The chain can then never be lost in a workup: you drive each reaction to completion with a flood of excess reagent, and simply wash the excess away. The demonstration molecule was a humble tetrapeptide, and the original chemistry is worth stating correctly because retellings usually get it wrong: Merrifield used Cbz protecting groups and DCC coupling. Boc chemistry arrived in 1967; the Fmoc group was introduced by Carpino and Han in 1970. Merrifield invented the solid phase — two generations of protecting-group chemistry were built on top of it afterwards (PMID 18429226).
The method's reach was proven in 1969 with the synthesis of ribonuclease A, a 124-residue enzyme, and in 1984 Merrifield received the Nobel Prize in Chemistry — alone, unshared — "for his development of methodology for chemical synthesis on a solid matrix" (history reviewed in PMID 18213693; Merrifield's own retrospective is PMID 3961484). Because the process is a mechanical loop, it was automated almost immediately, and every research peptide sold today descends from that loop.
Solid-phase peptide synthesis (SPPS) builds a chain backwards relative to biology — from the C-terminus toward the N-terminus — in a repeating four-step cycle (PMIDs 18429226, 29185032):
Side chains carry their own semi-permanent protecting groups throughout, because many amino acid side chains would otherwise react mid-synthesis. When the last residue is on, a single cleavage treatment cuts the finished chain off the resin and strips all side-chain protection at once. What comes off the bead is the crude product — the target peptide plus everything the process made alongside it.
The two classical regimes differ in how the temporary group comes off — and the difference decided which one the world uses.
Boc/Bzl (the original). The temporary Boc group is removed each cycle with moderate acid (classically ~50% TFA), while benzyl-type side-chain protection survives those cycles and is removed at the end by the strongest acid in the building: liquid anhydrous hydrogen fluoride, about an hour at −5 to 0 °C in a dedicated Teflon apparatus, with scavenger chemicals to catch the reactive fragments released (PMID 31879918). The structural weakness is baked in: temporary and permanent groups answer to the same kind of reagent, so every cycle's acid nibbles a little side-chain protection and a little peptide off the resin (PMID 26785684).
Fmoc/tBu (the regime that displaced it). The Fmoc group is removed not by acid but by base — standardly 20% piperidine in DMF — while the side-chain groups (tBu-type) ignore base entirely and come off in one final acid treatment. The scheme is truly orthogonal: two independent chemistries, no cross-talk, no per-cycle nibbling (PMID 18429226). Fmoc cleavage also releases a strongly UV-absorbing fragment, so the synthesizer can watch each deprotection happen and flag a failed step in real time (PMID 26785684).
The displacement was fast and it was measured: a survey of peptide synthesis core facilities found Fmoc use rising from 50 percent in 1991 to 98 percent in 1994 (PMID 18429226) — driven by avoiding HF, by easier automation, and by compatibility with acid-sensitive modifications like phosphorylation and glycosylation. Today Fmoc is the default for essentially all catalogue peptides (PMID 26785684). Boc survives honourably in niches: it remains the straightforward route to C-terminal thioesters (piperidine destroys them, and thioesters are needed for native chemical ligation) and for some long, difficult sequences.
Build your protocol, log every dose, monitor your body's response, and get reminders so you never miss a dose.
Start Tracking FreeThe pharmaceutical taxonomy of peptide impurities (D'Hondt et al. 2014, PMID 25044089) reads like a diary of the synthesis — each family is a specific step failing in a specific way. These impurities "possess the capability of greatly influencing initial functionality studies… possibly resulting in erroneous conclusions," which is why this section belongs in a buyer's guide and not just a chemistry text.
SPPS is a chain of near-identical steps, and step yields multiply. The table below is arithmetic, not measurement — it assumes each coupling succeeds in the stated fraction of chains and everything else is perfect, which is optimistic:
| Per-step yield | 10 steps | 20 steps | 40 steps | 50 steps |
|---|---|---|---|---|
| 99.9% | 99.0% | 98.0% | 96.1% | 95.1% |
| 99.0% | 90.4% | 81.8% | 66.9% | 60.5% |
| 98.0% | 81.7% | 66.8% | 44.6% | 36.4% |
Read the middle row. At 99 percent per step — performance most chemists would call excellent — a 40-coupling synthesis delivers a crude product in which only about two chains in three are the intended molecule. The rest is the near-relative family from the previous section, and purification must now remove it. Real syntheses are worse than the model in one way (deprotections and side reactions levy their own per-cycle taxes, and aggregation makes failures arrive in correlated runs, not independently) — the table is a ceiling argument, not a prediction for any product. But it explains, with no reference to any vendor's honesty, why crude purity must fall as length rises, why long peptides cost disproportionately more, and why "the same sequence" from two manufacturers is not automatically the same material.
Dosing cheat sheet, reconstitution reference, and cycle planning — delivered to your inbox.
Cleavage. One treatment cuts the chain from the resin and strips side-chain protection simultaneously. A worked Fmoc example from the primary literature: TFA with triisopropylsilane, phenol and water scavengers at 92.5 : 2.5 : 2.5 : 2.5 for 90 minutes (PMID 29307075). The scavengers exist because departing protecting groups are reactive carbocations that would otherwise re-attach to the peptide. Incomplete deprotection at this stage leaves protecting-group adducts — cleavage is a yield-and-purity step, not a formality (PMID 25044089).
Purification. Preparative reversed-phase HPLC is the predominant method (Chen, Mant and Hodges 2007, PMID 17156789), necessary precisely because synthesis impurities are close structural relatives of the target. Its governing trade is structural: collect a narrow slice of the main peak and purity rises while yield falls; collect wide and the reverse. In the one fully reported case in that study, the target was only 36.5 percent of the crude material, and a good semi-preparative protocol recovered about 91 to 93 percent of that fraction — one case, of a difficult antimicrobial peptide, quoted here as an illustration of the mechanics and not as an industry average. This trade is the mechanical origin of both grade and price: a "purified" product has paid for separation in discarded mass, time, and solvent; a "crude" product has not, and contains the full impurity family by design.
The powder. Purified fractions are freeze-dried (lyophilised) into the familiar white cake. Whatever counter-ion the peptide carried stays in the powder — and since TFA is present at cleavage and in the HPLC mobile phase, synthetic peptides are born as TFA salts. Acetate or hydrochloride forms require a deliberate extra exchange step (PMID 29307075); most approved peptide medicines are supplied as acetate salts (PMID 33287352). Two vials of equal gross weight but different salt forms do not contain equal amounts of peptide — which is why the companion guide treats net peptide content as the number that matters on a certificate.
What has improved. Industrial building blocks now exceed 99 percent purity themselves, removing a whole class of incorporated errors (PMID 26785684); microwave-assisted synthesis — precise, uniform heating rather than any exotic effect, per the independent critical review (PMID 22012213) — shortens syntheses and often raises crude purity, with the caveat that temperature-sensitive residues need adjusted conditions; and process research targets the wash steps that dominate SPPS waste (PMID 38071224, instrument-maker authors, claims attributed accordingly). None of it repeals the exponent in the table above. Every improvement pushes per-step yield toward 100 percent; none reaches it.
SPPS is one manufacturing route, not the definition of a peptide. Where length makes the arithmetic untenable, biology takes over: a gene encoding the product is installed in a production microbe and fermentation does the assembling, at ribosomal fidelity. The canonical case is insulin — 51 residues, two chains — produced recombinantly in E. coli and yeast since the first recombinant version in 1978 (PMID 25270715). One industry rule of thumb places the crossover at roughly 40 amino acids: chemical synthesis is standard below it, biological production typical above (PMID 38071224).
The two worlds also merge. Semaglutide's backbone is produced by fermentation and its defining lipid side chain is then installed chemically — semisynthesis, the general craft of assembling proteins from synthetic and recombinant fragments (PMID 31774265). Fragment strategies answer the arithmetic directly: make two clean 30-mers and join them — by native chemical ligation or enzymatic ligation — instead of one dirty 60-mer. More than 80 peptide drugs have reached the market across all these routes (PMID 33536635). The research-catalogue peptides in this library, though, are essentially all SPPS products, typically in the length range where the compounding table is the law they live under.
Nothing above tells you where to buy, because sourcing is the part that varies most. Whoever you use, require the same three things:
The vendor we recommend — full panel published per batch. A reader discount use it at checkout.
For research use only. Not for human consumption. 21+.
Build your protocol, log every dose, monitor your body's response, and get reminders so you never miss a dose.
Start Tracking FreeBruce Merrifield, at Rockefeller University, in 1963 (J Am Chem Soc 1963;85:2149–2154, doi:10.1021/ja00897a025). He received the 1984 Nobel Prize in Chemistry alone for it. Two common corrections: his 1963 paper used Cbz protecting groups and DCC coupling — Boc chemistry came in 1967 and Fmoc in 1970 — and the invention was the solid support itself, the idea that let excess reagents be washed away instead of purified away.
The temporary protecting group and its removal chemistry. Boc is removed each cycle with moderate acid, and final cleavage needs liquid hydrogen fluoride — and because both the temporary and permanent groups answer to acid, each cycle erodes the product slightly. Fmoc is removed with base (20% piperidine) while acid-labile side-chain groups wait for a single final acid step: a truly orthogonal scheme with no HF and machine-readable monitoring. Core facilities went from 50% Fmoc in 1991 to 98% in 1994 (PMID 18429226); Boc survives mainly for thioester production and some difficult sequences.
Multiplication. Each coupling succeeds on a fraction of chains, and fractions compound: at 99% per step, 40 steps leave about 67% of chains as the full-length target (0.99^40 ≈ 0.669) — and that model is optimistic, since deprotections, side reactions like aspartimide, and aggregation add their own per-cycle losses. The remaining third is a family of near-identical impurities that purification must remove, which is also why longer peptides cost more per milligram: inputs scale linearly while yield falls geometrically.
A peptide that finished synthesis missing one internal residue, because a single coupling or deprotection step did not go to completion on those chains and the synthesis simply continued without it. Deletions are among the most important synthesis impurities (PMID 25044089) precisely because they are near-copies of the target: one amino acid lighter, similar chromatographic behaviour, and documented to co-elute with the main peak in standard HPLC in at least one well-studied case (PMID 40857144, covered in the COA guide).
Because purification is where the money goes. Crude product is the cleaved, freeze-dried mixture — target plus its full family of deletions, truncations, epimers and adducts, legitimately useful for some purposes and priced accordingly. A purified grade has been through preparative HPLC, where narrow fraction collection buys purity at the cost of discarded mass and time (PMID 17156789). The grades describe process steps actually performed, not regulated categories — which is why a stated purity should always be read with its method, as the companion COA guide sets out.
Disclaimer: This guide is for educational and informational purposes only and is not intended as medical advice, diagnosis, or treatment. The compounds discussed are not FDA approved for human use. Always consult a qualified healthcare provider before starting any new supplement or peptide protocol. StackTrax does not sell peptides or supplements directly — purchase links go to third-party vendors. StackTrax is not responsible for the products, quality, or business practices of any third-party vendor.
© 2026 StackTrax, LLC. All rights reserved.
Privacy · Terms · Do Not Sell or Share My Personal Information
StackTrax guides cover peptides and compounds that are not FDA-approved for the uses discussed. The dosing, reconstitution, and safety information is compiled from published research and community protocols for educational purposes only.
Before using any compound mentioned here, consult a qualified healthcare provider. StackTrax does not sell, prescribe, or recommend these substances for personal use.