Mass Spectrometry Non Target Screening for Food Contact Degradation Products

Extract

A gas chromatography high resolution mass spectrometry run that detects twelve unidentified peaks above fifty micrograms per kilogram voids the supplier declaration of conformity for a food contact polyolefin laminate. That single outcome halts container release at the border. Under Article 3 of Regulation (EC) 1935/2004, materials contacting food cannot transfer constituents in quantities that endanger human health, bring about unacceptable changes in food composition, or cause deterioration of organoleptic characteristics.

When resins undergo compounding, extrusion, and lamination, thermal stress and shear induce polymer backbone cleavage and additive breakdown. These secondary chemicals are non-intentionally added substances. They inhabit the finished film without appearing on the raw material declaration or the positive list of Regulation (EU) 10/2011.

Laboratories quantify intentionally added starting monomers using reference standards and targeted tandem spectrometry. Non-target screening operates under inverted conditions: the analytical sequence encounters unknown chemical structures across a broad dynamic range without matched calibration curves. The screening workflow combines non-target high-resolution instrumentation with exact mass extraction, isotopic fidelity filtering, and software-driven database interrogation.

Liquid chromatography coupled to electrospray ionization quadrupole time of flight mass spectrometry captures polar, non-volatile migrants, while electron ionization high-resolution gas chromatography isolates volatile degradation fragments. High mass resolution above 30,000 full width at half maximum and sub-two-part-per-million mass accuracy distinguish genuine parent ions from matrix background noise.

A non-target feature remains unquantified until an authentic chemical standard matches both chromatographic retention time and tandem mass fragmentation patterns.

The regulatory evaluation of unassigned mass peaks rests on the Threshold of Toxicological Concern concept formalised by the European Food Safety Authority. Compounds lacking structural alerts for genotoxicity are evaluated against the Cramer Class III threshold of ninety micrograms per person per day. Translating this daily intake into food contact compliance assumes an adult consumes one kilogram of food contacting six square decimetres of packaging per day.

This exposure model establishes a screening threshold of ten micrograms per kilogram of food or food simulant. Any unassigned chromatographic peak generating an analytical response equivalent to ten micrograms per kilogram in food simulant D1 or simulant D2 triggers mandatory chemical identification and toxicological evaluation.

Semi-quantification introduces substantial uncertainty into this threshold evaluation. When an analyst estimates the concentration of an unidentified peak by assuming the response factor of an internal standard like deuterated benzophenone or 2,4,6-tribromophenol, the actual concentration may diverge by a factor of ten or higher. Ionisation efficiency in electrospray mass spectrometry varies across four orders of magnitude based on molecular structure, matrix suppression, and solvent composition.

An extraction yield that appears benign under a toluene-derived response factor can mask an actual migration exceeding the specific migration limit of an analogue chemical.

Pyrolysis

Thermal Stress and Additive Cleavage

Compounding polyolefins at temperatures between 190 and 260 degrees Celsius triggers oxidative cascades. Hindered phenol antioxidants, added to scavenge alkyl and peroxy radicals, sacrifice their own chemical structures during stabilization. Octadecyl 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionate degrades under processing shear and thermal exposure into 2,6-di-tert-butyl-1,4-benzoquinone, 3,5-di-tert-butyl-4-hydroxybenzaldehyde, and cinnamate derivatives.

These degradation products cross the phase boundary between the polymer bulk and the food simulant far more rapidly than the high molecular weight parent molecule.

Phosphite processing stabilizers present an identical analytical complication. Tris(2,4-di-tert-butylphenyl) phosphite oxidizes systematically during melt processing to yield tris(2,4-di-tert-butylphenyl) phosphate. Thermal scission of the aryl-phosphite ester linkage simultaneously liberates 2,4-di-tert-butylphenol.

That phenol fragment is a frequent migrant in liquid chromatography non-target screening profiles. The breakdown cascade proceeds according to the primary pathways documented in packaging qualification dossiers.

1. Primary Phenolic Hydrogen Donation produces an intermediate phenoxy radical that reacts further with alkylperoxy radicals to form peroxyquinones, which undergo thermal rearrangement to alkylated quinone methides and stable benzoquinones.
2. Secondary Hydroperoxide Decomposition by trivalent organic phosphites converts the antioxidant into its corresponding oxidized phosphate while suppressing autocatalytic polymer chain scission.
3. Hydrolytic Cleavage of Phosphite Esters initiated by ambient humidity during storage hydrolyzes the central phosphorus core into phosphorous acid, releasing free mono- and di-alkylated phenols into the polymer matrix.
4. Polymer Backbone Beta Scission generates terminal alkenes and short-chain volatile organic compounds including linear aldehydes and carboxylic acids, which shift the sensory profile of aqueous simulants.

Non-Intentionally Added Substances Cascade

Recycled post-consumer resins amplify chemical diversity. Mechanical recycling subjects previously processed polymers to repeated extrusion thermal cycles. Additive packages in recyclate contain heterogeneous combinations of legacy stabilizers, slip agents like erucamide, antistatic glycerol monostearates, and photo-oxidation fragments.

When secondary amines like hindered amine light stabilizers degrade, they yield nitroxy radicals, hydroxylamines, and low molecular weight alkylated amino compounds that lack entries in positive lists.

Polymer processing aids and slip agents also generate secondary transformation products through ambient oxidation. Erucamide undergoes allylic oxidation at its carbon-carbon double bond during corona treatment or blown film processing, yielding docosanamide epoxide and oxygenated fatty fragments. These degradation products alter the surface tension of the food packaging while introducing chromatographic peaks in both positive electrospray and electron impact ionization runs.

An importer receiving a declaration of conformity stating base polymer compliance remains legally accountable for these unlisted reaction byproducts.

Assay

Extracting degradation products from plastic packaging requires controlled contact conditions that preserve labile chemical structures. Simulant selection follows Annex III of Regulation (EU) 10/2011. Ten percent ethanol acts as simulant A for aqueous foods, three percent acetic acid represents simulant B for acidic matrices, and rectified olive oil, vegetable oil, or fifty percent ethanol serves as simulant D1 and D2 for lipophilic foods.

Polyolefin degradation testing frequently employs ninety-five percent ethanol or isooctane as substitute simulants for fat-containing items to avoid the non-volatile lipid background of natural oils, which suppresses ion transmission in mass spectrometers.

Standard migration testing applies exposure conditions defined in Annex V of Regulation (EU) 10/2011. Testing for long-term ambient storage requires migration cells containing simulant at forty degrees Celsius for ten days. For high-temperature processing like hot filling or retort sterilization, testing protocols escalate to two hours at seventy degrees Celsius or two hours at one hundred degrees Celsius followed by ambient contact.

Polymer samples undergo single-side immersion in certified stainless steel migration cells to prevent artificial extraction from the non-contact outer face.

Under EN 13130 migration conditions, ten days at forty degrees Celsius in ninety-five percent ethanol accelerates antioxidant ester hydrolysis beyond rates observed in actual aqueous food contact.

Gas chromatography paired with high-resolution mass spectrometry handles the volatile and semi-volatile chemical space below five hundred Daltons. Derivatization using BSTFA with one percent TMCS converts polar phenolic degradation products into volatile trimethylsilyl derivatives prior to gas chromatographic separation. Solid phase microextraction and purge-and-trap systems isolate volatile hydrocarbon oligomers and short-chain aldehydes directly from migration simulants, eliminating solvent evaporation steps that cause compound loss.

Liquid chromatography coupled to quadrupole time of flight or Orbitrap mass spectrometers resolves polar, thermolabile, and non-volatile migrants ranging from two hundred to twelve hundred Daltons. Reversed-phase C18 chromatography utilizing an acidified water and methanol gradient isolates polar degradation products under electrospray positive and negative ionization modes. Hydrophilic interaction liquid chromatography captures ultra-polar acidic oxidation products that co-elute with the void volume on standard reversed-phase columns.

Atmospheric pressure chemical ionization serves as an indispensable bridge between electrospray and electron ionization techniques. Compounds lacking acidic or basic functionalities fail to generate measurable ions in electrospray workflows. Atmospheric pressure chemical ionization ionizes neutral, intermediate-polarity antioxidant degradation products like alkylated quinones through gas-phase charge transfer reactions.

Deploying dual ionization platforms across both gas and liquid chromatography guarantees broad chemical coverage for non-target screening profiles.

High mass resolution and isotopic pattern evaluation form the basis of preliminary molecular formula attribution. Instruments operate at resolving powers exceeding 60,000 at m/z 200, achieving sub-one-millidalton mass accuracy. Mass spectrometers record both full scan precursor spectra and data-dependent or data-independent tandem mass spectra.

Collision energy stepping produces fragmentation patterns across low and high dissociation energies, capturing diagnostic product ions that confirm functional group substructures.

Triage

Data Reduction and Background Deconvolution

A typical high-resolution data acquisition file contains tens of thousands of individual ion features. Extracting relevant degradation products from this data stream requires rigorous chemometric subtraction. Chromatographic software deconvolutes overlapping isotopic peaks, aligns retention times across sample replicates, and performs background subtraction against method blank extracts.

The method blank isolates background contaminants introduced by procedural solvents, solid-phase extraction cartridges, chromatographic septa, and laboratory plastics.

Feature filtering proceeds through three successive mathematical steps: signal-to-noise ratio verification, replicate repeatability filtering, and blank thresholding. Features with peak areas less than five times the blank response are eliminated from the candidate list. Replicates must demonstrate chromatographic peak area relative standard deviations below twenty-five percent.

This filtration reduces raw feature counts from fifteen thousand detected masses to fewer than two hundred candidate migrants requiring identification.

Confidence Levels in Identification

Communicating non-target screening results requires a formal identification confidence scale. Packaging compliance workflows adopt the five-tier identification framework established in environmental and packaging analytical chemistry.

• Level 1 Confirmed Structure identifies a migrant by matching retention time, accurate mass precursor ion, and tandem mass fragmentation spectra with an authentic reference standard measured on the identical analytical platform under identical operating conditions.
• Level 2 Probable Structure assigns a single chemical identity based on matching high-resolution experimental tandem spectra against spectral databases like NIST, MassBank, or mzCloud, supported by chromatographic retention behavior.
• Level 3 Tentative Candidate designates a structure where accurate mass, isotopic fine structure, and diagnostic product ions establish a specific chemical family or positional isomer without distinguishing individual structural configurations.
• Level 4 Unequivocal Molecular Formula derives a unique elemental composition from high mass accuracy and isotopic distribution analysis when spectral databases contain no matching fragmentation spectra.
• Level 5 Exact Mass Feature records an unassigned monoisotopic mass and retention time without sufficient isotopic fidelity or fragmentation detail to define an elemental formula.

What Parameters Drive Identification Confidence?

The progression from Level 5 to Level 1 hinges on four analytical criteria: mass measurement error, isotopic pattern score, retention time index agreement, and diagnostic tandem fragment assignment. When resolving power drops below 30,000, isobaric interferences compromise mass measurement accuracy, producing false molecular formula assignments. Mass error must remain under two parts per million for compounds below four hundred Daltons.

Isotopic abundance ratios for carbon, nitrogen, sulfur, and chlorine must align with theoretical distributions within a five percent margin of error.

In-silico fragmentation engines like MetFrag and CFM-ID assist in evaluating Level 3 candidate structures. These computational tools apply rule-based dissociation models to candidate chemical structures imported from chemical databases such as PubChem or ChemSpider. The algorithms generate theoretical product ion fragments and assign matching scores against experimental tandem spectra.

Retention time prediction models based on quantitative structure-retention relationships provide an orthogonal validation filter to eliminate proposed chemical structures whose calculated partition coefficients conflict with observed retention times.

Scrutiny

The Threshold of Toxicological Concern serves as the primary toxicological arbiter for migrating non-intentionally added substances that lack empirical toxicological dossiers. Compounds identified at Level 2, 3, or 4 undergo structural evaluation using the Cramer decision tree. Software tools like Toxtree and the OECD QSAR Toolbox classify chemical structures into Cramer Classes I, II, or III based on chemical functionality, reactivity, and metabolic stability.

Cramer Class I covers simple chemical structures with low oral toxicity, corresponding to an intake threshold of 1800 micrograms per person per day. Cramer Class III comprises complex structures, reactive groups, or substances containing elements other than carbon, hydrogen, oxygen, nitrogen, and sulfur, restricted to ninety micrograms per person per day.

Converting the Cramer Class III threshold into an analytical screening limit requires applying conventional consumption assumptions. Article 17 of Regulation (EU) 10/2011 assumes an individual consumes one kilogram of packaged food daily. Dividing ninety micrograms by one kilogram of food yields an allowable migration concentration of ninety micrograms per kilogram.

For substances where structural alerts indicate potential genotoxicity, the applicable threshold drops to 0.15 micrograms per person per day, equivalent to 0.01 micrograms per kilogram of food or ten parts per billion. Analytical non-target screening methods cannot routinely achieve ten parts per billion limits of detection across uncalibrated chemical spaces. This limitation forces laboratories to establish the standard operational non-target screening limit at ten micrograms per kilogram, matching the general threshold applied to non-authorized substances behind functional barriers.

A supplier packaging guarantee omitting non-target screening data remains legally vulnerable to Article 3 enforcement actions when local market authorities detect unlisted migrants above ten parts per billion.

The critical vulnerability in non-target toxicological triage sits in the semi-quantification calculation. Because authentic calibration standards do not exist for unidentified degradation products at Level 3 and Level 4, concentrations are estimated using surrogate internal standards. The mathematical relationship governing semi-quantification relies on the ratio of the analyte peak area to the internal standard peak area multiplied by the known spike concentration.

Analytes possessing low electrospray ionisation efficiencies yield suppressed chromatographic signals relative to well-ionizing surrogate standards. A migrant whose true migration is fifty micrograms per kilogram can produce a chromatographic peak area corresponding to five micrograms per kilogram under a surrogate calibration curve based on benzophenone. This underestimation bypasses toxicological review and misrepresents legal compliance in the regulatory dossier.

To mitigate semi-quantification error, accredited laboratories deploy cocktail mixtures of internal standards spanning broad physicochemical spaces. The internal standard cocktail incorporates acidic, basic, neutral, polar, and non-polar deuterated compounds added at known concentrations prior to extraction. Quantifying an unidentified feature against the closest matching internal standard based on chromatographic retention time and polarity reduces response factor divergence.

This procedure tightens quantification error bands from two orders of magnitude down to a factor of two or three.

Verdict

Worked Exposure and Barrier Failure Assessment

Evaluating conformity requires translating analytical peak areas into legal verdicts. Consider a 50-micrometre multilayer film constructed with an outer polyethylene terephthalate skin, an intermediate polyurethane laminating adhesive, and an inner 25-micrometre linear low-density polyethylene food contact sealing layer. The package encloses 500 millilitres of an oil-in-water salad dressing emulsion containing fifteen percent vegetable lipids, corresponding to contact with food simulant D1.

The packaging contact area measures 3.5 square decimetres for 500 grams of food, establishing an actual surface-to-mass ratio of 7.0 square decimetres per kilogram, exceeding the conventional 6.0 square decimetre assumption.

The film undergoes migration testing under condition OM2 defined in Regulation (EU) 10/2011: ten days at forty degrees Celsius using simulant D1. High-resolution liquid chromatography Orbitrap screening under negative electrospray mode detects an unidentified peak at retention time 8.42 minutes with accurate mass m/z 219.1391 in negative ionization, corresponding to a deprotonated molecular ion of formula C14H20O2 with a mass error of 0.6 parts per million. Semi-quantification against a deuterated bisphenol A internal standard yields an apparent migration concentration of 14 micrograms per kilogram of simulant.

Applying the measured surface-to-mass ratio of 7.0 square decimetres per kilogram increases the corrected exposure concentration to 16.3 micrograms per kilogram of food. Database interrogation and in-silico fragmentation match the precursor mass to 3,5-di-tert-butyl-4-hydroxybenzaldehyde, a known oxidation product of phenolic antioxidant stabilizers. Cramer tree classification assigns this compound to Cramer Class III with a toxicological threshold of ninety micrograms per person per day.

Applying the operational ten microgram per kilogram screening limit categorizes this feature as an actionable non-intentionally added substance requiring Level 1 confirmation.

The regulatory and financial risk cascade is clear: when a brand places this article on the market without structural confirmation and an associated in-house toxicological dossier, border inspection authorities running routine market surveillance will flag the unlisted migrant, issue an analytical non-compliance report under Article 3 of Regulation (EC) 1935/2004, and initiate rapid alert border rejections that stall downstream logistics.