How to prove a lot is soy protein peptides
A soy protein peptide lot is identified by measurements that agree: the document names soy and the hydrolysis, protein by Kjeldahl, peptide content and distribution by HPLC, ash and moisture closing the mass balance, and a soy statement that matches the source. A protein percentage alone can be met by soy isolate, another plant protein or non-protein nitrogen, so it proves neither the hydrolysis nor the source.
Identity is the agreement of the source statement, the peptide distribution, the nitrogen figure and the mass balance; a protein percentage alone proves none of them.
The five identity layers
A name on a certificate is only a claim until a measurement backs it. Each layer below asks a different question and needs its own result; a supplier who supplies only a protein figure has not described the powder.
| Layer | What it asks | Method | What it establishes |
|---|---|---|---|
| Document | Are soy and the hydrolysis named? | Name, source, process description | Paperwork only |
| Protein | How much nitrogenous material is present? | Kjeldahl with the factor stated | Quantity, not the source |
| Peptide | Is the protein actually hydrolysed? | HPLC distribution with the cut-off stated | The peptide fraction |
| Mass balance | Is the powder diluted? | Moisture, ash and amino acid profile | Consistency of the numbers |
| Allergen | Is soy declared? | Source-based allergen statement | The label obligation |
The layers are independent. A lot can be genuine soy and still be diluted; it can pass protein and peptide and still carry another plant protein; it can be the right source and the wrong hydrolysis. Each needs its own evidence, and the evidence belongs in the batch file rather than in a marketing sentence.
The peptide-to-protein ratio and the ash ratio
The simplest screen is the ratio between the two headline figures. At 90.0% protein and 80.0% peptide on the same basis, the peptide fraction is 88.9% of the protein; the remaining 11.1 points are free amino acids, peptides below the HPLC cut-off and non-protein nitrogen. A wider gap points to a different hydrolysis, a different method or an addition.
The ratio also catches the common substitutions. Adding free amino acids raises the Kjeldahl figure while the peptide figure stays flat, so the ratio falls. Selling soy isolate as a peptide raises protein and leaves the peptide fraction low, so the ratio falls further. Diluting with a carrier lowers both figures, so the ratio can look normal while the absolute numbers drop.
Ash is a second ratio. At ≤5.0% ash and ≥90.0% protein, the ash-to-protein ratio is at most 0.056. A lot with 90% protein and 8% ash carries either a mineral load or a neutralisation salt that the specification did not intend. Moisture and ash must be converted to the same basis before the mass balance is closed; a dry-basis protein figure added to an as-is moisture figure will not add up.
Where XRD and FTIR stop
Spray-dried soy peptide has no crystal lattice, so an X-ray diffraction scan returns a broad halo instead of sharp peaks. That is what an amorphous hydrolysate should look like, but the halo says nothing about soy and nothing about hydrolysis. Sharp reflections point to something crystalline: a free amino acid, an inorganic salt, a mineral carrier or sugar that crystallised in storage.
XRD is not part of the routine release panel here. When a supplier reports “XRD conforms”, ask for the pattern, the reference and the question the test was meant to answer. FTIR is quicker: amide I near 1650 cm⁻¹ and amide II near 1545 cm⁻¹ show the peptide backbone. It confirms the material class, not the source, and cannot separate soy from pea or wheat.
What adulteration looks like
| Route | What it changes | Signal | Test |
|---|---|---|---|
| Soy isolate or another intact protein | Protein high, peptide low | Peptide-to-protein ratio low; opaque dispersion | Peptide HPLC; solubility |
| Another plant protein | Protein passes, source is not soy | Amino acid profile shift; soy marker absent | Amino acid profile; PCR for soy |
| Free amino acids | Protein up, peptide flat | Peptide-to-protein ratio low | Peptide HPLC; amino acid profile |
| Non-protein nitrogen | Kjeldahl up, peptide flat | Amino acid sum short of the protein figure | Amino acid profile; melamine by LC-MS/MS |
| Maltodextrin or carrier | Both headline figures down | Ash, moisture and mass balance shift | Mass balance; sugar HPLC |
A PCR test for soy targets DNA, and a highly purified protein can carry little of it, so a negative result supports the file rather than proving absence on its own. The allergen declaration does not depend on the PCR result.
Colour proves the least. Off-white to light yellow is a wide window, and a carrier can be milled to match it. The peptide distribution, the amino acid profile and the mass balance carry the real evidence. For soy there is a second control: the allergen statement. A lot sold as soy peptide must carry the soy declaration, and an offer to remove it because the material is hydrolysed gets the obligation backwards.
How to write identity into the specification
- “Hydrolysed soy protein, sold as soy protein peptides, with the soy source named.”
- “Protein: ≥90.0% by Kjeldahl; the factor and the moisture basis appear on the COA.”
- “Peptide: ≥80.0% by HPLC; column, mobile phase and cut-off named.”
- “Moisture ≤7.0%, ash ≤5.0%, ≥95% passing 80 mesh.”
- “Amino acid profile and molecular-weight distribution available on request.”
- “No added plant protein, free amino acids, non-protein nitrogen or carrier.”
- “Soy allergen statement; non-GMO or identity-preserved status only where claimed.”
Five questions for the supplier
- Which soy material went into the hydrolysis, and what enzyme, time and temperature were used?
- What factor and moisture basis sit behind the protein figure?
- Which HPLC method and cut-off produced the peptide figure, and can you send the distribution?
- What is the amino acid profile, and how does its sum compare with Kjeldahl?
- What does the soy allergen statement say, and will you support a joint re-test on the retained sample?
Five direct answers amount to an identity file. A single protein percentage does not. The product hub carries the catalog grade; the COA page shows where each figure sits on the certificate; the grades page compares the peptide and the isolate; the level-four comparison covers the label consequences; the methods section explains what each assay proves.
Related documents
Send the specification you intend to use — grade, quantity and destination market — and we will reply with the batch COA, the technical data sheet and a quotation.
Request a quotation