Mineral and Energy Economy Research Institute of the Polish Academy of Sciences

The dataset contains the characterisation data of four commercial biomass carbonizates, the input data and code of their multi-criteria screening as activated-carbon precursors, and the fixed-bed adsorption data of decamethylcyclopentasiloxane (D5) on a steam-activated carbon produced at industrial scale from the selected carbonizate. The carbonizates are coconut shell, standard grade (NR1); coconut shell, black grade (NR2); palm kernel shell (NR3); and almond shell (NR4). The activated carbon (product name COCO-1000) was obtained from NR1 by steam activation in an industrial installation (TRL 9).

The data are organised in five folders packed in one ZIP archive.

1. 01_Carbonizates: ash content and volatile matter (triplicates); SEM micrographs (TIFF), EDS map spectra (EMSA) and quantification reports with and without oxygen (RTF), with all quantification results in one CSV table; XRF reports of the ash (PDF, three replicates per sample) with replicate and mean tables; XRD patterns (Rigaku RASX and CSV); and raw thermogravimetric data of CO2 gasification at 930 °C (NETZSCH export and UTF-8 CSV), with a Python script that calculates conversion curves, t50, t90 and the reactivity R = 1/t50.
2. 02_MCDA: the decision matrix (EDS carbon purity, ash content, EDS inorganic fraction and CO2 reactivity) and a Python script that ranks the carbonizates by the weighted-sum model (WSM) and TOPSIS under fixed weighting scenarios and in a Monte Carlo analysis of 20,000 weight vectors (Dirichlet distribution, fixed random seed), with the resulting scores, rank-1 probabilities and mean ranks.
3. 03_Activated_carbon: quality parameters of the activated carbon (iodine number, methylene blue number, ash, moisture, mechanical strength, bulk density; three measurements each) and its XRD pattern.
4. 04_D5_breakthrough: breakthrough data (Cout/Cin at 1 h intervals over 10 h) for four particle fractions (2.5–4, 1–2, 0.5–1 and 0.25–0.5 mm) in three independent runs each, measured in dry nitrogen at an inlet D5 concentration of 300 mg/m3 with a sorbent mass of 1.00 g; the experimental conditions; and a Python script that calculates breakthrough and saturation times, total capacity, removal efficiency, bed utilisation at breakthrough and Yoon–Nelson and Thomas model parameters.
5. 05_Figure_scripts: Python scripts that reproduce the figures of the related article from the data above.

Instruments used: Axia ChemiSEM LV scanning electron microscope with TrueSight X EDS detector (Thermo Fisher Scientific), ScopeX 980CS XRF analyser, SmartLab SE diffractometer (Rigaku) and STA 449 F3 Jupiter thermal analyser (NETZSCH); D5 concentrations were determined by TD-GC-MS. Tabular data are provided as UTF-8 CSV files with a comma separator and a decimal point; the scripts require Python 3.9 or newer with numpy, scipy and matplotlib. Abbreviations: EDS – energy-dispersive X-ray spectroscopy; MCDA – multi-criteria decision analysis; PKS – palm kernel shell; SEM – scanning electron microscopy; TD-GC-MS – thermal desorption gas chromatography–mass spectrometry; TOPSIS – Technique for Order Preference by Similarity to Ideal Solution; WSM – weighted-sum model; XRD – X-ray diffraction; XRF – X-ray fluorescence. For a detailed description of all files, measurement conditions and column definitions please consult the readme.txt file.