Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC 61198: Determination of 2-Furfural in Mineral Insulating Oils โ Transformer Paper Aging Assessment
💡
IEC 61198 is the internationally recognized analytical standard specifying methods for the determination of 2-furfural (2-FAL) and related furanic compounds in mineral insulating oils. It provides the chemical diagnostic basis for non-invasive assessment of cellulose paper insulation degradation in power transformers — arguably the single most important parameter for estimating remaining transformer life.
1. Technical Fundamentals and Standard Scope
The operational life of oil-filled power transformers is ultimately limited by the condition of their solid insulation system — cellulose paper and pressboard. Cellulose, a linear polymer composed of anhydroglucose rings linked by β-1,4-glycosidic bonds, undergoes chain scission under the combined effects of thermal stress, moisture, and oxygen. This depolymerization process releases characteristic furanic compounds into the insulating oil, chief among them being 2-furfuraldehyde (2-FAL), which accounts for 70-90% of total furanic compounds generated during normal service aging.
First published in 1992 and most recently revised as IEC 61198:2023, the standard establishes two analytical methods: a reference High-Performance Liquid Chromatography (HPLC) method for accurate quantitative analysis, and a UV-VIS spectrophotometric screening method for field use. The standard’s critical contribution to the industry lies in providing a standardized, reproducible protocol that ensures cross-laboratory comparability of results, enabling utilities worldwide to build consistent transformer condition assessment programs.
✅
Core Chemical Principle: Cellulose degradation produces five principal furanic compounds: 2-furfural (2-FAL), 5-hydroxymethyl-2-furfural (5-HMF), 2-acetylfuran (2-ACF), 5-methyl-2-furfural (5-MEF), and 2-furoic acid (2-FCA). Among these, 2-FAL is the predominant and most thermally stable species, exhibiting the strongest correlation with the average degree of polymerization (DP) of paper insulation.
2. Analytical Methods — Principles and Practical Implementation
2.1 HPLC Reference Method
The HPLC method specified by IEC 61198 is the definitive analytical procedure, offering high sensitivity (LOD ~0.01 mg/L), excellent selectivity, and robust reproducibility. The analytical workflow comprises three sequential stages:
Liquid-Liquid Extraction: A measured volume of oil sample (typically 10 mL) is vigorously shaken with an equal volume of acetonitrile. Furanic compounds partition preferentially into the acetonitrile phase due to their higher polarity relative to the hydrocarbon oil matrix. Single-stage extraction efficiency typically ranges from 85% to 95% depending on the specific furanic species and oil type. For highest accuracy, a second extraction step may be performed, and the combined extracts are evaporated and reconstituted to a known volume.
Chromatographic Separation: The acetonitrile extract is injected into a reversed-phase HPLC system equipped with a C18 bonded silica column (typically 250 mm × 4.6 mm, 5 μm particle size). Isocratic or gradient elution using acetonitrile-water mixtures achieves baseline separation of all five target compounds within 15-25 minutes. The elution order is governed by compound polarity: 5-HMF elutes first (most polar), followed by 2-FCA, 2-FAL, 2-ACF, and finally 5-MEF (least polar).
UV Detection and Quantification: Detection is performed at 280 nm, near the absorption maximum of 2-FAL. Quantification uses the external standard method by comparing peak areas against a calibration curve constructed from reference standards at five or more concentration levels. The use of an internal standard (e.g., 2-ethylfuran or 2-propylfuran) is strongly recommended to compensate for extraction efficiency variations and injection volume fluctuations.
| Furanic Compound | Abbreviation | λ_max (nm) | Relative Retention Time | Typical Range (mg/L) | Aging Significance |
|---|---|---|---|---|---|
| 2-Furfural | 2-FAL | 278 | 1.00 (reference) | 0.01 – 10 | Primary degradation indicator |
| 5-Hydroxymethyl-2-furfural | 5-HMF | 284 | 0.65 | 0.005 – 1 | Early-stage aging marker |
| 2-Acetylfuran | 2-ACF | 272 | 1.20 | 0.005 – 0.5 | Minor degradation product |
| 5-Methyl-2-furfural | 5-MEF | 292 | 1.45 | 0.005 – 2 | Advanced aging marker |
| 2-Furoic Acid | 2-FCA | 254 | 0.50 | 0.01 – 1 | Oxidative degradation product |
2.2 UV-VIS Spectrophotometric Screening Method
The UV-VIS method provides a simplified, equipment-minimal approach suitable for on-site preliminary screening. Direct absorbance measurement of the oil sample (or its acetonitrile extract) at 280 nm is correlated to 2-FAL concentration via a pre-established calibration curve. While operationally convenient, this method suffers from significant matrix interferences: oil oxidation byproducts, antioxidant additives (BHT, DBPC), and other aromatic compounds all absorb in the same UV region, leading to potential overestimation. The method is therefore classified as semi-quantitative and is suitable only for screening purposes.
⚠️
Engineering Recommendation: UV-VIS screening is acceptable for routine monitoring of normally aged transformers. However, when 2-FAL concentration exceeds 0.1 mg/L, when critical maintenance decisions are pending, or when the transformer operating history shows anomalies, HPLC confirmation is mandatory. Relying solely on UV-VIS data for replacement or overhaul decisions carries substantial risk of either unnecessary intervention or missed critical degradation.
2.3 Critical Sampling and Handling Considerations
Sample integrity is paramount. The following engineering practices, distilled from IEC 61198 requirements and field experience, are essential for reliable results:
Sampling Valve and Flushing: Always collect oil from the bottom sampling valve after flushing 1-2 liters of oil to eliminate stagnant oil from the dead volume. For transformers without dedicated sampling valves, use a clean stainless steel sampling device.
Container Specifications: Use amber glass bottles (to prevent UV-induced photodegradation of furanic compounds) with PTFE-lined caps. The container must be completely filled to eliminate headspace, minimizing volatilization losses and oxidation during storage.
Preservation and Transport: Samples must be protected from light at all times. Transport temperature should be maintained at 4-10°C, and analysis should be completed within 48 hours of sampling. For long-term storage (beyond 7 days), freezing at -18°C or below is required. Avoid plastic containers — furanic compounds can adsorb onto polyethylene surfaces, causing significant concentration losses.
2.4 Method Validation and Quality Assurance
IEC 61198 specifies method performance criteria that laboratories must demonstrate for accreditation. Key validation parameters include: repeatability (RSD ≤ 5% for HPLC, ≤ 15% for UV-VIS), reproducibility (RSD ≤ 15% across laboratories), recovery (85-115% for spiked samples), and linearity (r² ≥ 0.998 over the calibration range). Participation in inter-laboratory proficiency testing programs (e.g., those organized by CIGRE or national standardization bodies) is strongly recommended to ensure ongoing analytical competence.
3. Engineering Application — 2-FAL in Transformer Life Assessment
3.1 Condition Classification Based on 2-FAL Concentration
Extensive field data from thousands of power transformers worldwide has enabled the establishment of practical condition classification thresholds. The following table summarizes the generally accepted interpretation guidelines:
| 2-FAL Concentration (mg/L) | Insulation Condition | Recommended Action |
|---|---|---|
| < 0.1 | Normal aging | Routine annual monitoring |
| 0.1 – 0.5 | Mild degradation | Reduce monitoring interval to 6 months |
| 0.5 – 1.0 | Moderate degradation | Add DGA and DP testing; evaluate remedial options |
| 1.0 – 5.0 | Severe degradation | Plan replacement or major refurbishment |
| > 5.0 | End-of-life condition | Immediate evaluation; consider decommissioning |
3.2 Quantitative Correlation with Degree of Polymerization
The relationship between oil 2-FAL concentration and paper DP has been extensively studied. Among the various proposed models, the semi-logarithmic correlation originally developed by Chendong and colleagues remains the most widely applied in engineering practice:
log₁₀(2-FAL, mg/L) = 1.51 – 0.0035 × DP
This model predicts that as DP declines from an initial value of 1000-1200 to approximately 250 (the universally accepted end-of-life criterion for cellulose insulation), 2-FAL concentration rises from approximately 0.01 mg/L to 5-10 mg/L. It is crucial to recognize that this relationship is influenced by several factors including transformer operating temperature profile, moisture content of the paper, oil type (naphthenic vs. paraffinic), and the transformer’s breathing system design (free-breathing vs. nitrogen-blanketed).
🔴
Critical Engineering Caveat: A single 2-FAL measurement must never be interpreted in isolation. Key confounding factors include: (1) Oil treatment or replacement removes furanic compounds without improving paper condition; (2) Free-breathing transformers lose furanic compounds through volatilization, leading to underestimation; (3) Thermal siphoning effects can create non-uniform distribution of 2-FAL within the transformer. Always correlate 2-FAL data with DGA results (particularly CO/CO₂ ratios), electrical tests (capacitance, dissipation factor), and operating history for a reliable assessment.
3.3 Trend Analysis — The Gold Standard for Interpretation
In the authors’ experience spanning over two decades of transformer condition assessment, the temporal trend of 2-FAL concentration provides far more diagnostic value than any single measurement. Establishing a baseline measurement within the first year of commercial operation, followed by annual or semi-annual tracking, enables calculation of the aging rate (d[2-FAL]/dt).
A particularly sensitive indicator is the concentration doubling time. Under normal operating conditions at typical load cycles, the 2-FAL doubling time ranges from 3 to 8 years. A significant reduction in doubling time (e.g., from 5 years to 18 months) is an early warning of accelerated aging requiring immediate investigation — often traceable to increased moisture ingress, sustained overloading, or compromised cooling system performance.
4. Practical Engineering Recommendations and Future Outlook
💡
Summary of Best Practices:
① Establish a 2-FAL baseline for every transformer within 12 months of commissioning.
② Conduct annual 2-FAL testing as the minimum frequency; adjust based on trend.
③ Always combine 2-FAL data with DGA, insulation resistance, and dielectric response measurements.
④ Use HPLC as the primary analytical method; reserve UV-VIS for field screening only.
⑤ Focus on trend analysis — the rate of change matters more than the absolute value.
⑥ Maintain a centralized database for fleet-wide comparison and anomaly detection.
① Establish a 2-FAL baseline for every transformer within 12 months of commissioning.
② Conduct annual 2-FAL testing as the minimum frequency; adjust based on trend.
③ Always combine 2-FAL data with DGA, insulation resistance, and dielectric response measurements.
④ Use HPLC as the primary analytical method; reserve UV-VIS for field screening only.
⑤ Focus on trend analysis — the rate of change matters more than the absolute value.
⑥ Maintain a centralized database for fleet-wide comparison and anomaly detection.
The emerging landscape of online sensor technology promises real-time 2-FAL monitoring. Optical fiber-based sensors employing UV absorption spectroscopy and electrochemical sensors with molecularly imprinted polymers have demonstrated laboratory feasibility. However, current field-proven solutions remain limited by drift, cross-sensitivity, and long-term stability challenges. It is expected that future revisions of IEC 61198 may incorporate provisions for validated online monitoring methods, aligning with the broader industry trend toward predictive maintenance and digital substation concepts.
5. Frequently Asked Questions (FAQ)
❓ Q: Can a very low 2-FAL concentration (below 0.01 mg/L) guarantee that paper insulation is in excellent condition?
Q: Not necessarily. Low 2-FAL can result from: (a) the transformer truly being in early service life with DP above 800; (b) the oil having been replaced or reconditioned, removing accumulated furanic compounds; (c) a free-breathing design where furanic compounds continuously volatilize; or (d) the transformer operating at consistently low temperatures where degradation is minimal. Cross-check with CO/CO₂ ratios from DGA and, if access is available, direct DP measurement of paper samples from the tap changer or pump outlet.
❓ Q: What are the main sources of error in HPLC determination of 2-FAL?
Q: The principal error sources include: (1) Variable extraction efficiency due to oil matrix effects — different base oils and additive packages affect the partition coefficient of furanic compounds into acetonitrile; (2) Sample degradation during storage — photodegradation and volatilization are the primary mechanisms; (3) Chromatographic interferences — co-extracted oil components such as phenolic antioxidants (BHT, DBPC) may produce UV absorption at 280 nm, potentially causing false positives; (4) Standard stability — 2-FAL reference standards are susceptible to oxidation and should be verified periodically. Using internal standard quantification and regular participation in proficiency testing programs are the most effective mitigation strategies.
❓ Q: How should 2-FAL data be interpreted in combination with DGA results?
Q: The combination of 2-FAL and DGA provides powerful diagnostic synergy. High 2-FAL with elevated CO and CO₂ (especially a CO/CO₂ ratio above 0.33) confirms cellulose involvement in the aging process. High 2-FAL with normal DGA suggests uniform thermal aging without active faults. Low 2-FAL with high H₂ and C₂H₂ suggests partial discharge or arcing faults that have not significantly affected the paper yet. Low 2-FAL with high CO/CO₂ may indicate that paper degradation products have been flushed out by oil replacement. A comprehensive multi-parameter assessment always outperforms single-indicator interpretation.
❓ Q: Does the type of insulating oil (naphthenic vs. paraffinic, uninhibited vs. inhibited) affect 2-FAL analysis?
Q: Yes, to varying degrees. Naphthenic oils generally exhibit higher solubility for furanic compounds than paraffinic oils, which can affect the partition equilibrium during liquid-liquid extraction. The presence of antioxidant additives (BHT at typical concentrations of 0.3-0.8% by mass) does not significantly interfere with HPLC analysis but may cause substantial positive bias in UV-VIS measurements. The IEC 61198 HPLC method has been validated across the major oil types used in power transformers globally, but method validation is recommended when establishing testing capability for a new oil formulation or supplier.
