Volatile matter determination is a fundamental test for coal, coke, and solid biofuels, and a rigorous evaluation of its measurement uncertainty is essential for reliable quality assessment and compliance with international standards. This article provides a comprehensive technical overview of the uncertainty evaluation process, covering the measurement model, major uncertainty sources, quantification methods, and reporting requirements in accordance with the GUM framework and recognized test standards such as ISO 562, ASTM D3175, and GB/T 212. It serves as a practical reference for laboratory professionals, quality engineers, and anyone seeking to understand the reliability of volatile matter test results.
Technical Overview
Volatile matter (VM) is one of the most important quality parameters for solid fuels, including coal, coke, and biomass pellets. It is defined as the mass loss of a test sample, other than moisture, produced when the sample is heated under strictly standardized conditions, typically at 900 °C ± 10 °C for 7 minutes in a covered crucible with restricted air access. Because the result is highly sensitive to experimental conditions, the credibility of a volatile matter value cannot be judged from a single number alone; it must be accompanied by a quantified measurement uncertainty.
Measurement uncertainty evaluation for volatile matter determination follows the general framework described in ISO/IEC Guide 98-3 (the Guide to the Expression of Uncertainty in Measurement, GUM) and, where repeatability data from quality control charts or interlaboratory studies are available, ISO 21748. The core procedure involves four steps: constructing a measurement model, identifying and quantifying all significant uncertainty sources, calculating the combined standard uncertainty using the law of propagation of uncertainty, and reporting the expanded uncertainty, typically at a coverage factor of k = 2 corresponding to a confidence level of approximately 95%.
For volatile matter determined on an air-dried basis, the simplified measurement model is expressed as the percentage mass loss during heating minus the moisture content of the sample. Each input quantity in this model — initial sample mass, mass after heating, and moisture content — contributes its own uncertainty component, which must be combined with the repeatability of the whole procedure to obtain a realistic overall estimate. Accredited third-party laboratories performing this test under ISO/IEC 17025 accreditation are required to estimate and report uncertainty as part of their technical competence.
Testing Samples
Coal and Coal Products
Bituminous coal, anthracite, sub-bituminous coal, and lignite represent the largest category of samples. Coal is heterogeneous by nature, so sample preparation (crushing, division, and air-drying to less than 0.2 mm particle size) introduces variability that directly affects the repeatability component of uncertainty. High-volatile coals generally exhibit larger relative mass losses and therefore require careful control of crucible handling and weighing.
Coke and Carbonized Materials
Metallurgical coke, foundry coke, and semi-coke are characterized by low volatile matter contents, typically below a few percent. For low-level results, fixed uncertainty contributions such as balance resolution and moisture correction become proportionally more significant, which increases the relative uncertainty of the reported value.
Solid Biofuels and Biomass Pellets
Wood pellets, agricultural residues, and other compressed biomass are tested under the specific heating conditions of ISO 18123 for solid biofuels. Biomass samples often contain higher and more variable moisture levels, making the moisture determination term a dominant contributor to the overall uncertainty budget.
Other Carbonaceous Materials
Charcoal, activated carbon precursors, and various carbon materials can also be characterized for volatile content using adapted gravimetric procedures, provided that the heating temperature, duration, and atmosphere are defined in the test protocol and reflected in the measurement model.
Testing Items
The evaluation of uncertainty in volatile matter determination typically covers the following reported quantities:
- Volatile matter content on air-dried basis (Vad), dry basis (Vd), and dry-ash-free basis (Vdaf)
- Combined standard uncertainty of the volatile matter result (uc)
- Expanded uncertainty (U) at a coverage factor of k = 2, corresponding to approximately 95% confidence
- Uncertainty of the auxiliary moisture determination used for basis conversion and correction
- Repeatability and reproducibility statistics obtained from replicate measurements and control charts
- Uncertainty contributions from mass measurements, furnace temperature, heating time, and sample preparation
Testing Methods
Standardized Determination Procedures
The gravimetric determination is performed according to recognized standards, most commonly ISO 562 for hard coal and coke, ASTM D3175 for coal and coke, GB/T 212 for Chinese commercial coal analysis, and ISO 18123 for solid biofuels. A pre-weighed covered crucible containing approximately 1 g of analysis sample is placed into a muffle furnace already stabilized at 900 °C ± 10 °C, held for exactly 7 minutes (or the standard-specified duration), cooled in a desiccator, and reweighed. The volatile matter is calculated from the mass loss with appropriate moisture and, where required, decomposition corrections.
Identification of Uncertainty Sources
A systematic uncertainty analysis identifies the following principal components:
- Repeatability (Type A evaluation): the scatter of replicate volatile matter determinations on the same or similar samples, usually the single largest contributor
- Balance resolution and calibration: the readability and calibration uncertainty of the analytical balance applied to both the initial and final weighings
- Furnace temperature: deviation from and fluctuation around 900 °C, plus temperature uniformity within the crucible zone, since mass loss is exponentially temperature-dependent
- Heating time: timing deviations in placement, removal, and recovery of the furnace temperature after sample insertion
- Moisture determination: the uncertainty of the moisture value subtracted in the calculation, propagated directly into the volatile matter result
- Sample representativeness and preparation: particle size distribution, air-drying equilibrium, and sample heterogeneity
- Crucible and handling effects: lid sealing integrity, cooling conditions, and possible oxidation or mass gain of low-volatile samples
Calculation and Combination of Uncertainty
Each quantified source is converted into a standard uncertainty (for example, dividing a rectangular distribution by the square root of three where full bound corrections apply). Sensitivity coefficients are derived from the measurement model, typically partial derivatives of volatile matter with respect to sample mass, residual mass, and moisture. The combined standard uncertainty is obtained as the root-sum-square of all contributing components. The expanded uncertainty is then calculated as U = k × uc with k = 2, and the final result is reported in the form V ± U, for example 28.4 % ± 0.6 % (k = 2).
Verification of the Uncertainty Budget
The estimated uncertainty is validated by comparison with the repeatability and reproducibility limits published in the test standards, by participation in proficiency testing schemes among accredited third-party laboratories, and by ongoing statistical control of routine QC samples. If observed between-replicate variation exceeds the budgeted repeatability term, the evaluation must be revised, as this indicates an uncontrolled influence quantity.
Applications
Uncertainty evaluation for volatile matter determination supports a wide range of industrial and regulatory applications:
- Coal trade and contract settlement: demonstrating whether a consignment meets specification limits with stated confidence, including pass or fail decisions against tolerance boundaries
- Power generation: boiler and burner design, combustion stability assessment, and fuel blending decisions that depend on reliable volatile matter values
- Coking and metallurgy: classification of coking coals and prediction of coke yield and quality
- Biomass energy: fuel specification verification for wood pellets under EN ISO solid biofuel standards and international trade documentation
- Laboratory accreditation: fulfilling the ISO/IEC 17025 requirement to estimate uncertainty for relevant test methods and to consider it in statements of conformity
- Method validation and comparison: assessing the equivalence of results from different furnaces, operators, or test standards
- Regulatory compliance: supporting customs classification, emission-related fuel declarations, and certification programs
FAQ
What is volatile matter and why does it need an uncertainty statement?
Volatile matter is the mass loss, excluding moisture, released when a sample is heated under defined standardized conditions, mainly consisting of combustible gases such as hydrogen, methane, carbon monoxide, and hydrocarbon tars. Because the result depends on precise temperature, time, and mass measurements, an uncertainty statement quantifies the range within which the true value is expected to lie, giving users a defensible basis for decisions.
What are the main sources of uncertainty in volatile matter testing?
The dominant contributors are usually the repeatability of the whole procedure and the furnace temperature control, followed by the uncertainties of the initial and final mass measurements, the heating time, the auxiliary moisture determination, and the representativeness of the prepared analysis sample.
How is the expanded uncertainty calculated and reported?
All standard uncertainty components are combined by the root-sum-square method using sensitivity coefficients from the measurement model, yielding the combined standard uncertainty. This is multiplied by a coverage factor of k = 2 to give the expanded uncertainty U, which corresponds to approximately 95% confidence, and the result is reported as V ± U with a statement of the coverage factor.
What is a typical expanded uncertainty for coal volatile matter?
For medium-to-high volatile bituminous coals tested by experienced laboratories, expanded uncertainties on the order of 0.3 % to 0.7 % absolute (k = 2) are commonly achieved, provided the furnace and balance are well controlled. Low-volatile materials such as coke can show larger relative uncertainty because fixed contributions weigh more heavily at low result levels.
How can laboratories reduce the uncertainty of volatile matter results?
Effective measures include strict furnace temperature calibration and zoning verification, precise timing protocols, calibrated balances with adequate resolution, consistent crucible handling and desiccator cooling, careful sample preparation to the specified particle size, regular moisture cross-checking, and statistical process control using certified reference materials or routine QC samples.
Does the reported basis of the result affect the uncertainty?
Yes. Converting from an air-dried basis to a dry or dry-ash-free basis introduces the uncertainties of the moisture and ash determinations through additional propagation terms, so the expanded uncertainty of Vdaf is generally larger than that of the directly measured air-dried value.