Trace determination of ethylene glycol (EG) residual content is a critical quality and safety test for pharmaceutical excipients, polyesters and food-contact packaging, cosmetics, and ethylene oxide–sterilized medical devices. This technical article reviews the regulatory limits, typical sample types, and validated analytical strategies—including headspace GC-FID, derivatization GC-MS, and LC-MS/MS—for accurately quantifying ethylene glycol at ppm to ppb levels.
Technical Overview
Ethylene glycol (ethane-1,2-diol, CAS 107-21-1) is a highly polar, hygroscopic diol with a boiling point of 197.3 °C. It enters commercial products as a residual monomer in polyesters such as PET, as a process solvent or impurity in pharmaceutical manufacture, as a co-product in glycerin and propylene glycol supply chains, and as a hydrolysis by-product of ethylene oxide sterilization. Because EG is metabolized in vivo to glycolate and oxalate—causing central nervous system depression, metabolic acidosis, and acute kidney injury—its residual content is strictly controlled.
Key regulatory benchmarks include ICH Q3C(R8), which classifies ethylene glycol as a Class 2 residual solvent with a permitted daily exposure (PDE) of 6.2 mg/day and a concentration limit of 620 ppm in drug substances and drug products. In food-contact materials, EU Regulation 10/2011 assigns ethylene glycol and diethylene glycol a combined specific migration limit of 30 mg/kg food. Pharmacopoeial monographs for glycerin typically restrict EG and DEG to not more than 0.1%, and ISO 10993-7 governs residues associated with ethylene oxide sterilization of medical devices, where EG is commonly quantified alongside ethylene oxide and ethylene chlorohydrin.
Trace-level EG measurement is analytically demanding for several reasons:
- Low volatility and a high boiling point limit sensitivity in conventional static headspace GC.
- Strong polarity and hydrogen-bonding capacity cause peak tailing and adsorption on active sites in GC systems.
- The molecule lacks a UV chromophore, making HPLC-UV unsuitable without derivatization.
- Complex matrices—glycerin, propylene glycol, polyethylene glycols, and DEG—introduce co-elution and matrix effects that must be resolved chromatographically or by mass-selective detection.
Testing Samples
Pharmaceutical Ingredients, Excipients, and Drug Products
Glycerin, liquid polyethylene glycols (PEG 200–600), polysorbates, and propylene glycol are screened for EG as an impurity arising from ethylene oxide chemistry. Oral solutions, syrups, suspensions, and drug substances processed with glycol-based solvents are also submitted for residual EG verification and batch release.
Polymer and Packaging Materials
PET granules, preforms, bottles, films, sheets, trays, and fibers are tested for residual ethylene glycol monomer. Polyester polyols, alkyd resins, powder coatings, and thermoset laminates used in food-contact and pharmaceutical packaging are analyzed both as materials and in migration extracts.
Medical Devices
Ethylene oxide–sterilized products—PVC tubing, catheters, infusion sets, syringes, wound dressings, surgical drapes, and procedure kits—are extracted and analyzed for EG as part of sterilization residue evaluation under the ISO 10993 framework.
Cosmetics and Personal Care Products
Glycerin-rich moisturizers, toothpastes, mouthwashes, serums, and PEG-based formulations are tested for EG to support safety assessments under regulations such as EU Regulation 1223/2009.
Industrial and Specialty Matrices
Polyester-grade monoethylene glycol streams, textile fibers, gel coats, resin systems, deicing fluids, and forensic or investigational samples are analyzed for residual or contaminant EG at trace levels.
Testing Items
Testing is tailored to the matrix and the applicable specification:
- Residual ethylene glycol content, reported as mg/kg (ppm, w/w).
- Individual and combined quantification of ethylene glycol and diethylene glycol (EG + DEG).
- EG in food-simulant migration extracts and device extraction media.
- Method development, validation, and transfer support in accordance with ICH Q2(R2) and USP general chapter requirements.
- Batch release testing, stability programs, out-of-specification investigations, and supplier qualification screening.
Testing Methods
Method selection depends on the expected concentration, the sample matrix, and the required reporting limit. A tiered strategy combines routine GC-FID screening with mass-selective confirmation for trace-level work.
Sample Preparation
- Liquid samples are weighed and diluted in ultrapure water or methanol, with internal standard addition (typically 1,3-propanediol, 1,4-butanediol, or isotopically labeled EG-d4 for MS detection).
- Polymer solids are cryogenically ground and extracted by sonication at 60 °C, accelerated solvent extraction, or direct thermal desorption in the headspace vial at 100–140 °C using DMSO as a dispersing solvent.
- Food-contact migration extracts are prepared in 10% and 50% ethanol, 3% acetic acid, or isooctane simulants under standard oven conditions.
- Saturated sodium chloride is added to headspace vials to improve EG partitioning into the vapor phase; standard addition is used to compensate for the strong matrix effects of glycerin-rich samples.
Static Headspace GC-FID
Samples are equilibrated for 30–45 minutes at 90–120 °C in sealed vials and analyzed on a polyethylene glycol (wax) column, for example 30 m × 0.32 mm × 0.25 µm, with flame ionization detection. The wax phase fully resolves EG from propylene glycol and DEG. This approach is robust and cost-effective for routine quality control with quantitation limits of approximately 5–10 ppm.
Direct-Injection GC-FID
For liquid matrices at concentrations above roughly 50 ppm, direct split or splitless injection with a deactivated retention guard minimizes discrimination and active-site adsorption. This method is useful for rapid screening but carries a higher risk of inlet contamination from non-volatile matrix components.
Derivatization GC-MS
For trace-level determination, EG is converted to its bis(trimethylsilyl) ether using BSTFA with 1% TMCS in dry pyridine or acetonitrile at 70 °C for 30 minutes. The derivative is analyzed on a low-bleed 5%-phenyl methylpolysiloxane column with SIM acquisition, monitoring m/z 73 and 147 as characteristic fragments and the molecular ion at m/z 206. This strategy removes volatility limitations and achieves quantitation limits of approximately 0.2–1 ppm with unambiguous mass-spectral confirmation.
LC-MS/MS and Complementary Techniques
Benzoyl chloride derivatization followed by reversed-phase LC-MS/MS with multiple-reaction monitoring enables LOQs down to 0.01–0.1 ppm, and is the method of choice for glycerin- and propylene-glycol-dominated matrices, since separation relies on derivatized analyte hydrophobicity rather than volatility. High-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) offers an orthogonal option that separates EG, DEG, and glycerol without derivatization, while classical periodate oxidation–spectrophotometric assays remain useful for screening. HPLC with refractive index detection is reserved for higher-concentration QC applications.
Method Validation and Data Quality
Methods are validated per ICH Q2(R2) and USP <1225>. Typical acceptance criteria include: resolution of at least 1.5 from DEG, propylene glycol, and glycerin; linearity with r of 0.995 or better from the LOQ to 150% of the specification; mean recovery of 85–115% at three spiking levels; repeatability RSD of no more than 10% (15% at the LOQ); LOD and LOQ established at signal-to-noise ratios of 3 and 10; and demonstrated robustness to ±5 °C equilibration temperature, ±10 minutes equilibration time, and split-ratio variation. Procedural blanks, matrix blanks, and carryover checks (below 10% of the LOQ response) are included in every sequence because EG is ubiquitous in laboratory environments.
Applications
Trace EG determination directly supports product safety, regulatory compliance, and process control across multiple industries:
- Pharmacopoeial batch release of glycerin and propylene glycol against EG/DEG limits, and supply-chain due-diligence screening following historical diethylene glycol contamination incidents.
- Compliance dossiers for PET packaging and food-contact articles, demonstrating conformity with the combined EG/DEG specific migration limit of 30 mg/kg under EU Regulation 10/2011.
- Release testing of ethylene oxide–sterilized medical devices, quantifying EG together with ethylene oxide and ethylene chlorohydrin per ISO 10993-7.
- Polymer and resin process monitoring, where residual EG tracks polycondensation completeness in PET and polyester production.
- Cosmetic safety assessments, technical file support, and consumer-product investigations.
- Research and development support, including extractables and leachables studies and root-cause investigations for out-of-specification results.
FAQ
What detection limits can routinely be achieved?
Headspace GC-FID typically achieves LOQs of 5–10 ppm. Derivatization GC-MS reaches approximately 0.2–1 ppm, and derivatized LC-MS/MS can quantify EG down to 0.01–0.1 ppm, comfortably below the 620 ppm ICH Q3C limit and pharmacopoeial specifications.
Why is ethylene glycol difficult to analyze by standard GC methods?
EG has a high boiling point, low vapor pressure, and strong polarity, which suppress headspace sensitivity and cause adsorption and peak tailing on active GC surfaces. Elevated equilibration temperatures, salting-out, deactivated flow paths, and derivatization are used to overcome these limitations.
How much sample is required for testing?
Typical submissions are 0.5–2 g of solid material or 1–5 mL of liquid, supplied in tightly sealed containers. Samples are analyzed promptly to prevent moisture-related changes and environmental contamination.
How is EG distinguished from diethylene glycol and glycerin?
Wax-phase capillary columns baseline-resolve EG (bp 197 °C), propylene glycol (bp 188 °C), and DEG (bp 245 °C), while glycerin (bp 290 °C) is essentially non-volatile under headspace conditions. GC-MS or LC-MS/MS ion ratios provide orthogonal identity confirmation.
Are the methods compliant with pharmacopoeial and regulatory expectations?
Yes. Although EG is not efficiently recovered by the general static headspace procedure of USP <467> due to its low volatility, dedicated validated methods are developed and qualified against ICH Q2(R2) and USP <1225> principles, and reporting is aligned with ICH Q3C, EU 10/2011, and ISO 10993-7 requirements.
What is the typical turnaround time?
Standard testing is completed within 5–10 business days, including sequence-specific blanks, calibration, and QA review. Expedited 48–72 hour services are available for urgent release or investigation scenarios.