Nitrogen Spiking in Protein Raw Materials: How an Analytical Testing Laboratory Detects What a COA Misses

The Kjeldahl method has been the industry standard for protein quantification since 1883. That’s not a typo — the same chemistry developed during the Austro-Hungarian Empire is still the basis of most protein COAs your suppliers hand you today. And the operations selling adulterated protein raw materials know exactly how it works.

Nitrogen spiking — adding cheap, high-nitrogen compounds to artificially inflate apparent protein content — isn’t new. What’s changed is the sophistication and scale. A 2023 independent survey of 134 protein supplement finished goods found that 28% contained non-protein nitrogen sources not declared on the label. Behind many of those products is a raw material COA that passed incoming quality review without ever being sent to an independent analytical testing laboratory for verification.

If your receiving protocol ends at “compare results to supplier COA,” you’re not verifying protein content. You’re trusting arithmetic that was never designed to distinguish the real thing from a $1.00/kg filler.

Why the Kjeldahl and Dumas Methods Are Structurally Blind to Spiking

Both Kjeldahl (wet acid digestion) and Dumas (high-temperature combustion) measure total nitrogen in a sample, then multiply by a conversion factor — 6.25 for most mixed proteins — to estimate protein content. Neither method has any mechanism for distinguishing protein-bound nitrogen from nitrogen in free amino acids, creatine, taurine, urea, or, infamously, melamine.

Spikers exploit this with deliberate precision. Glycine costs roughly $0.80–$1.20 per kilogram wholesale, compared to $4.50–$7.00/kg for a quality whey protein concentrate. A 5% glycine addition to a protein blend raises the apparent nitrogen content enough to represent a protein content boost of approximately 3–4 percentage points on a Kjeldahl readout. At a 5,000 kg purchase order, that’s a significant margin advantage — funded entirely by your quality gap.

The amino acids most commonly identified in spiking operations aren’t chosen randomly. Glycine, taurine, creatine, beta-alanine, and L-alanine were selected because they carry high nitrogen-to-molecular-weight ratios and don’t trigger sensory or solubility red flags at low-to-moderate concentrations. Creatine, at roughly 32% nitrogen by weight, is particularly effective — it barely changes the texture or flavor profile of a whey concentrate even at 3–4% addition levels.

Your supplier’s COA reading “85.2% protein by Kjeldahl” can be technically accurate and practically useless in the same moment.

What a Qualified Analytical Testing Laboratory Actually Runs to Catch It

Detecting nitrogen spiking requires moving away from bulk nitrogen measurement and toward the molecular composition of the protein itself. A qualified analytical testing laboratory has several tools at its disposal, and the right strategy uses them in layers.

Amino Acid Profile by HPLC — AOAC 2016.01

This is the gold standard for protein adulteration detection. AOAC 2016.01 quantifies 17–20 individual amino acids following complete acid hydrolysis of the protein matrix. Because hydrolysis breaks all peptide bonds before analysis, you get a comprehensive picture of the protein’s actual composition. And every legitimate protein source has a predictable fingerprint. Whey protein concentrate will consistently show leucine at approximately 9.5–11%, isoleucine and valine together at 10–12%, and glutamic acid as the dominant single amino acid at 17–20%. If glycine shows up at 9–12% in a whey sample, that’s not lot-to-lot variation. That’s adulteration.

Full amino acid profiling typically requires 24–48 hours of analytical time, which makes it impractical as a screen for every incoming pallet. But for any new supplier, any first shipment from a returning supplier after a gap, or any lot where the nitrogen content is unexpectedly high relative to historical data — it’s non-negotiable.

Free vs. Bound Amino Acid Ratio Testing

A more targeted, faster approach. Spiked amino acids are added in free form; legitimate protein-derived amino acids exist in peptide-bound form. By running two HPLC extractions — one without hydrolysis (measuring free amino acids only) and one with hydrolysis (measuring total amino acids) — a lab can determine what fraction of the nitrogen source is protein versus filler.

In practice, free amino acid content above 3–5% of total protein content in a concentrate is anomalous enough to require explanation. Some manufacturers now hardcode this into their specifications: WPC-80, for instance, is often specced at ≤4% free amino acids as a percentage of total protein content. An isolate should sit closer to ≤2%.

NIR Spectroscopy as a Screening Gate

Near-infrared spectroscopy won’t give you a definitive confirmation, but a well-calibrated NIR model built against known-authentic reference standards can flag outliers in under 60 seconds per sample — fast enough to screen every incoming lot before any chemistry is run. Against a comprehensive whey protein reference library, a calibrated NIR instrument can catch gross spiking events with 85–90% sensitivity. We use it as a gate: samples that match the reference fingerprint proceed to COA comparison; samples that flag go directly to HPLC.

Isotope Ratio Mass Spectrometry (IRMS) for the Sophisticated Cases

For spiking involving amino acids that are chemically identical to their naturally-occurring counterparts — and therefore invisible to composition-based analysis — IRMS can distinguish synthetic from biosynthetically-derived amino acids based on their carbon-13 to carbon-12 isotope ratios. Synthetic glycine and natural glycine are compositionally identical but isotopically distinct. IRMS is expensive, with per-sample costs typically running $150–$300, and turnaround measured in days. But for high-value ingredients, suspected systematic fraud, or a regulatory dispute involving a major supplier, it’s definitive in a way that no other method is.

What Spiking Actually Costs You When It Slips Through

The immediate loss is straightforward: you paid for 80% protein and received 73% functional protein. On a 1,000 kg order, that’s a formulation deficit that either forces you to over-include the ingredient (costing more) or under-deliver on the label claim (costing credibility and regulatory standing).

The downstream exposure is harder to price but considerably more serious.

Under 21 CFR 111.75, dietary supplement manufacturers must establish and verify that incoming raw materials conform to established specifications before use in a finished product. If your specification requires only “≥80% protein by Kjeldahl” and your test confirms that — but the actual protein content is inflated by non-protein nitrogen — you’ve passed a test that didn’t measure what you needed to know. FDA will not credit you for a passing result on a method that missed the adulteration. The obligation to have specifications adequate to catch the failure sits with the manufacturer.

The FTC has taken enforcement action against supplement brands for protein content and efficacy claims that couldn’t be substantiated — and several of those cases trace back to raw materials with nitrogen-inflated COAs that passed internal testing. Between 2018 and 2022, at least two U.S. class-action settlements involved plaintiffs demonstrating that finished products with conforming nitrogen-based COAs contained measurably less bioavailable protein than the label claimed.

There’s also the formulation science problem. If your product depends on achieving a specific leucine threshold — the 2.5–3.0 g leucine per serving that the research literature consistently associates with maximal muscle protein synthesis signaling — you can’t get there from a whey protein where 8% of the amino acid content is glycine added by a supplier. The nitrogen numbers add up. The clinical outcomes don’t.

Building a Specification That Actually Stops Spiking at the Door

Most raw material specifications in the supplement industry were written to reflect what a supplier is capable of providing, not what a buyer scientifically requires. That’s a structural problem, and it’s one that spiking operations are built to exploit.

A protein raw material specification that genuinely protects you includes several elements beyond total protein by Kjeldahl:

• Protein content confirmed by amino acid profile (AOAC 2016.01) — either as the primary method or as a required qualifier when Kjeldahl is used as the primary screen
• Minimum limits for key functional amino acids — for WPC-80, this typically means leucine ≥9.5%, combined BCAAs ≥20%, glycine ≤3.5%, and creatine not detected at a defined LOQ
• Free amino acid content limit — ≤4% of total protein for concentrates, ≤2% for isolates, with any exceedance triggering a full AOAC 2016.01 profile
• Non-protein nitrogen (NPN) limit — expressible directly or achieved indirectly through the amino acid limits above
• Identity confirmation — at minimum a FTIR or NIR fingerprint match against an authenticated reference; for plant-based proteins, species confirmation via DNA or HPTLC is worth adding

Critically, verification must be performed by an independent analytical testing laboratory — not the supplier’s in-house quality team — on at least the first shipment from any new or re-qualified supplier, and on a risk-informed frequency thereafter. A supplier who has shipped clean for 18 months still has economic incentives that can shift. Periodic independent testing is what keeps those incentives honest.

The One Change That Catches Most of It

If a full specification overhaul isn’t feasible right now, there’s a single addition that provides most of the protection: a free amino acid screen by HPLC without hydrolysis, run alongside your standard nitrogen test on every protein lot.

The incremental analytical cost is typically $40–$80 per sample. Any free amino acid total exceeding 5% of stated protein content triggers a full AOAC 2016.01 profile before the lot is accepted. That’s it. That protocol, applied consistently, would have flagged the majority of documented nitrogen spiking cases in the industry over the past decade. It doesn’t catch everything — a sophisticated operation that pre-bonds its filler amino acids can sometimes pass a free amino acid screen — but it catches the economically-motivated, operationally-simple adulteration that accounts for the vast majority of cases.

The economics of nitrogen spiking depend entirely on your not looking closely enough. Run the right tests, set the right limits, and you stop being a target worth the trouble.

---

Written by Nour Abochama, VP Operations, Qalitex | Quality Consultant, Ayah Labs. Learn more about our team

Talk to our team about raw material testing. Contact us

Related from our network

• Protein Potency and Label Accuracy Testing for Supplement Brands — Qalitex Laboratories offers ISO 17025-accredited amino acid profiling and potency verification for finished supplement products and raw ingredients.
• How to Evaluate a Certificate of Analysis Before Accepting a Supplement Ingredient — Qalitex’s quality team breaks down what a COA should contain — and the limits of what it can actually guarantee.