Precision testing for the determination of fixed carbon content verifies that repeated measurements of coal, coke, graphite, and other carbonaceous materials produce statistically consistent results under repeatability and reproducibility conditions. This technical article explains the calculation basis of fixed carbon, the precision evaluation protocol aligned with ISO 17247, ASTM D3172, and GB/T 212, and the statistical criteria used by accredited third-party laboratories to demonstrate method control.
Technical Overview
Fixed carbon (FC) is the calculated, non-volatile combustible residue of a solid fuel or carbon material, obtained by difference after moisture (M), ash (A), and volatile matter (VM) have been determined on the same analytical basis: FC (%) = 100 − M − A − VM. Because the fixed carbon value is derived rather than measured directly, its uncertainty combines the uncertainties of the three component determinations, which makes a formal precision study an essential element of method validation.
Precision testing quantifies the closeness of agreement between repeated determinations under two defined conditions. Repeatability conditions involve the same operator, the same apparatus, and the same laboratory within a short time interval, while reproducibility conditions involve different operators, different apparatus, or different laboratories. The results are expressed as repeatability standard deviation (sr), reproducibility standard deviation (sR), relative standard deviation (RSD), repeatability limit (r = 2.8 × sr), and reproducibility limit (R = 2.8 × sR).
Testing programs are designed and evaluated in accordance with ISO 5725 (accuracy of measurement methods and results), ISO/IEC 17025 competence requirements, and the normative precision statements of product standards such as ISO 17247, ASTM D3172, ASTM D5142, and GB/T 212. Certified reference materials and statistical control charts are used to confirm that the determination remains in a state of statistical control over time.
Testing Samples
Coal
Bituminous coal, sub-bituminous coal, anthracite, thermal coal, and coking coal are the most frequently tested materials. Samples are air-dried, crushed, and ground to a nominal test-particle size of approximately 212 µm (0.2 mm) before analysis, ensuring homogeneity so that observed variability reflects the method rather than sample preparation.
Coke and Petroleum Coke
Metallurgical coke, foundry coke, green petroleum coke, and calcined petroleum coke exhibit high fixed carbon values, often above 85 percent on a dry basis. Precision testing at these high levels is particularly important because small relative errors in ash or volatile matter propagate into the fixed carbon result.
Graphite and Carbon Materials
Natural graphite, synthetic graphite, graphite electrode materials, carbon anodes, and carbon raisers are tested to verify consistency between production lots. These materials are typically low in volatile matter, so precision is dominated by moisture and ash determination.
Activated Carbon, Biochar, and Charcoal
Granular and powdered activated carbon, biochar, and lump charcoal are evaluated for fixed carbon content as an indicator of carbonization degree and adsorption performance. Precision testing supports process control in activation kilns and pyrolysis units.
Testing Items
A complete precision testing program for fixed carbon content typically includes the following measurable and derived quantities:
- Fixed carbon content on an air-dried basis (FC, ad) and on a dry basis (FC, d)
- Component determinations: total moisture and analytical moisture, ash yield, and volatile matter yield
- Repeatability standard deviation (sr) and repeatability limit (r) of the fixed carbon determination
- Reproducibility standard deviation (sR) and reproducibility limit (R) where interlaboratory data are available
- Relative standard deviation (RSD, coefficient of variation) at each concentration level tested
- Bias verification against certified reference materials or standard samples
- Optional expanded measurement uncertainty estimation consistent with the ISO/IEC Guide 98-3 (GUM) framework
Testing Methods
Gravimetric Proximate Analysis (Reference Method)
The classical approach follows ASTM D3173, ASTM D3174, and ASTM D3175 (under the umbrella practice of ASTM D3172), ISO 562, ISO 1171, ISO 11722, and GB/T 212. A test portion of approximately 1 g is sequentially heated: moisture is determined at 104–110 °C, ash at the standard ashing temperature (700–750 °C per ASTM, 815 °C per ISO and GB/T), and volatile matter at 900 ± 10 °C for seven minutes in a covered crucible with restricted air access. Fixed carbon is then calculated by difference on the reporting basis specified by the client.
Instrumental Macro-Thermogravimetric Method
Automated thermogravimetric analyzers determine moisture, ash, and volatile matter in a single programmed run with controlled atmosphere switching, following ASTM D7582, ASTM D5142, and ISO 17246. Instrumental methods offer higher throughput and reduced operator variability, but their precision must be verified against the gravimetric reference method during validation and at regular intervals thereafter.
Precision Evaluation Protocol
The precision experiment is structured according to ISO 5725-2 and generally comprises the following steps:
- Selection of at least three sample levels spanning the expected working range of fixed carbon content, for example approximately 45–60 percent, 60–80 percent, and above 80 percent
- Performance of a minimum of six to ten independent replicate determinations per level under repeatability conditions, with each replicate prepared and analyzed as a complete, separate run
- Optional interlaboratory study with multiple participating laboratories and at least two replicates per cell to establish reproducibility statistics
- Screening of data for outliers using Grubbs and Cochran tests at the 5 percent significance level before calculating pooled statistics
- Calculation of mean values, standard deviations, RSD, and the limits r and R, followed by comparison against the precision statements of the governing standard
Acceptance Criteria and Interpretation
Absence of a fixed, universal limit means the acceptance criteria must be taken from the applicable product or method standard for the specific material and concentration level. As general guidance, repeatability limits for fixed carbon in coal and coke proximate analysis are commonly on the order of 0.3 to 1.0 percentage points absolute, while reproducibility limits may extend to approximately 1.5 to 2.5 percentage points depending on the level. Results exceeding the applicable limit indicate the need to investigate sample preparation, furnace temperature calibration, balance performance, crucible geometry, and operator technique before reporting.
Applications
Precision testing of fixed carbon content determination supports quality control and commercial decision-making across multiple industries:
- Coal mining and trading: lot conformity assessment, price settlement based on calorific and compositional specifications, and coal classification according to international codification systems
- Coke and ironmaking: blast furnace charge quality control, CSR/CRI test programs, and calcined coke acceptance for anode production
- Carbon products manufacturing: consistency verification for graphite electrodes, carbon anodes, carbon raisers, and lubricating graphite powders
- Foundry operations: incoming inspection of foundry coke and recarburizers where fixed carbon directly affects melt chemistry
- Water treatment and air purification: qualification of activated carbon lots for carbonization degree and adsorption-related properties
- Laboratory accreditation and method validation: generation of precision data required for ISO/IEC 17025 scope extensions, internal quality control charts, and participation in proficiency testing schemes
FAQ
Why is fixed carbon determined by difference rather than measured directly?
No direct physical separation fully isolates the fixed carbon fraction. The proximate analysis scheme defines it as the residue remaining after moisture, ash, and volatile matter are removed, so calculation by difference is the internationally accepted convention. This also means the precision of fixed carbon depends on, and is always somewhat lower than, the precision of the best-determined component parameter.
How many replicates are required for a valid precision study?
Under ISO 5725, a minimum of six replicate determinations per level is generally accepted for evaluating within-laboratory repeatability, while interlaboratory reproducibility studies require at least eight participating laboratories with duplicate results each. More replicates improve the confidence interval of the estimated standard deviation.
What is the difference between repeatability and reproducibility?
Repeatability describes the agreement of results obtained by the same operator on the same equipment in the same laboratory within a short period. Reproducibility describes the agreement of results obtained in different laboratories or under changed conditions. Both are expressed as limits r and R, within which two single results should agree with 95 percent probability when the method operates correctly.
Which factors most commonly degrade the precision of fixed carbon results?
Frequent causes include non-homogeneous sample fineness, incomplete air drying before weighing, incorrect furnace temperature or ramp profile, improperly lidded volatile matter crucibles, insufficient balance resolution, and calculation of results on inconsistent moisture bases. A structured precision study identifies which step contributes the largest variance.
How often should precision be re-verified?
Accredited third-party laboratories typically verify precision at fixed intervals through quality control samples plotted on control charts, alongside periodic checks against certified reference materials and participation in interlaboratory proficiency testing. A full precision study is repeated whenever the method, instrument, crucible type, or sample matrix changes significantly.