Filtern
Dokumenttyp
Sprache
- Englisch (3)
Referierte Publikation
- ja (3)
Schlagworte
- Biodiesel (1)
- Concrete (1)
- Fuel (1)
- Hackathon (1)
- Inverse Design (1)
- Large Language model (1)
- Orchestration (1)
- Prediction (1)
- Reference measurement (1)
- Sulphur (1)
Organisationseinheit der BAM
The CCQM-K123 key comparison was organized by the Inorganic Analysis Working Group (IAWG) of CCQM to assess and document the capabilities of the national metrology institutes (NMIs) or the designated institutes (DIs) to measure the mass fractions of sodium, calcium, potassium, magnesium phosphorous and sulfur in biodiesel fuel (BDF). The National Metrology Institute of Japan (NMIJ) and National Institute of Standards and Technology (NIST) acted as the coordinating laboratories.
Results were submitted by 11 NMIs and DIs. The participants used different measurement methods, though most of them used inductively coupled plasma-mass spectrometry (ICP-MS), isotope dilution technique with ICP-MS and inductively coupled plasma-optical emission spectrometry (ICP-OES) with microwave acid digestion.
The material was quite challenging and a number of questions were raised at the IAWG meeting. Concerning S, the variation in S results between participants, particularly those using IDMS methods was discussed at the IAWG meeting. BAM, NIST and NMIJ reviewed their experimental conditions, results and/or uncertainty calculations for IDMS. According to the additional evaluation and investigation, the variances between the revised results became smaller than the original one, the revised results were overlapping between IDMS measurements of S content at the k=2 level. It is not possible to calculate a KCRV with values being modified after submission. It was concluded that this KC does not support S measurements.
Accounting for relative expanded uncertainty, comparability of measurement results for each of Na, Ca, K, Mg and P was successfully demonstrated by the participating NMIs or DIs. It is expected that sodium, calcium, potassium, magnesium and phosphorus at mass fractions greater than approximately 0.1 mg/kg, 0.1 mg/kg, 0.05 mg/kg, 0.05 mg/kg and 0.1 mg/kg respectively in biodiesel fuel and similar matrices (fuels and oils etc.) can be determined by each participant using the same technique(s) employed for this key comparison to achieve similar uncertainties mentioned in the present report. Furthermore, the results of this key comparison can be utilized along with the IAWG core capability approach.
Under the auspices of the Organic Analysis Working Group (OAWG) of the Comité Consultatif pour la Quantité de Matière (CCQM) a key comparison, CCQM K55.a, was coordinated by the Bureau International des Poids et Mesures (BIPM) in 2009/2010. Eleven national measurement institutes and the BIPM participated. Participants were required to assign the mass fraction of estradiol present as the main component in the comparison sample (CCQM-K55.a) which consisted of a bulk estradiol hemihydrate material obtained from a commercial supplier that had been extensively but not exhaustively dried prior to sub-division into the units supplied for the comparison.
Estradiol was selected to be representative of the performance of a laboratory's measurement capability for the purity assignment of organic compounds of medium structural complexity [molar mass range 300500 Da] and low polarity (pKOW < -2) for which related structure impurities can be quantified by capillary gas phase chromatography (GC) or by high performance liquid chromatography (LC).
The majority of participants used a mass balance approach to determine the estradiol content. The key comparison reference value (KCRV) for estradiol in CCQM-K55.a was assigned by combination of KCRVs assigned by consensus from participant results for each orthogonal impurity class. This allowed participants to demonstrate the efficacy (or otherwise) of their implementation of the mass balance approach and to demonstrate that their assigned value for the main component agreed with the KCRV through use of internally consistent contributing methods.
The KCRV for the estradiol content of the material was 984.3 mg/g with a combined standard uncertainty of 0.42 mg/g. The individual participant results showed that a relative expanded uncertainty for the purity assignment of 0.2% is a reasonable estimate of the best achievable result by an individual laboratory for a material of this complexity available in this amount at this level of purity. The relative expanded uncertainties reported by laboratories having results consistent with the KCRV ranged from 0.2% to 0.8%.
The review of results that were biased from the KCRV showed that two major analytical challenges are posed by the material: the measurement of its water content and controlling for related substance artefact formation during the analysis process. The results displaying a positive bias relative to the KCRV (overestimation of estradiol content) were due to underestimation of the water content of the material, while those with a negative bias (underestimation of estradiol) overestimated the total related substance impurities through a failure to detect and control for artefact formation arising from in situ oxidative dimerization of estradiol in neutral solution prior to analysis. There was however good agreement between all participants in the identification and the quantification of the individual related structure impurities actually present in the sample.
The comparison also demonstrated the utility of high-field 1H NMR for both quantitative and qualitative analysis of high purity compounds. It is noted that all the participants who used qNMR as a major or contributing technique and included it as part of, combined it or confirmed it with a conventional 'mass balance' data estimate, obtained results consistent with the KCRV.
Large-language models (LLMs) such as GPT-4 caught the interest of many scientists. Recent studies suggested that these models could be useful in chemistry and materials science. To explore these possibilities, we organized a hackathon. This article chronicles the projects built as part of this hackathon. Participants employed LLMs for various applications, including predicting properties of molecules and materials, designing novel interfaces for tools, extracting knowledge from unstructured data, and developing new educational applications. The diverse topics and the fact that working prototypes could be generated in less than two days highlight that LLMs will profoundly impact the future of our fields. The rich collection of ideas and projects also indicates that the applications of LLMs are not limited to materials science and chemistry but offer potential benefits to a wide range of scientific disciplines.