Factors Affecting the Accuracy and Reliability of the Analysis Results of the Carbon Sulfur Determination

At present, Carbon Sulfur Determination are generally used for the analysis of carbon and sulfur elements in metals. In the process of using analytical instruments of various manufacturers, there are often unsatisfactory repeatability or reproducibility. In this regard, the first thing to suspect is the failure of the instrument, but often after checking the hardware of the instrument, it is concluded that the instrument is working normally. In the following, for the analysis of the principle, operation and analysis process of the carbon and sulfur analyzer, several factors that affect the accuracy and reliability of the analysis results of the Carbon Sulfur Determination are summarized.

The amount of cosolvent added

Another factor that affects the stability of analytical results is the amount of co-solvent added, which is very prominent when analyzing low-level samples. For example, when analyzing samples with carbon and sulfur mass fractions less than 15 μg/g, 1.5 g and 2 g of co-solvent were added respectively (the carbon and sulfur mass fractions in the co-solvents were ≤8 μg/g, ≤5 μg/g, respectively), because the amount of co-solvent added is usually It does not participate in the calculation of the analysis results, so the fluctuation of the carbon and sulfur content in 0.5g of the co-solvent is introduced between the two analyses. Assuming that the content of carbon and sulfur in the co-solvent is 3 μg/g, and the sample weight is 0.5 g, a deviation of 3 μg/g is introduced due to the different amount of the co-solvent added.

Therefore, when analyzing low-content samples, it is necessary to ensure that the amount of co-solvent added is consistent, thereby reducing the uncertain factors introduced by the co-solvent.

Sample weighing

Different samples weighed different amounts of Carbon and Sulfur contained in them, resulting in different areas where the analysis results fall on the calibration curve of the instrument. Due to the limited linear range of the instrument, the difference in the correction area will cause fluctuations in the analysis results, especially when the upper and lower limits of the analytical instrument are near the impact.

For low-content samples, when the sample is weighed larger, the result will be higher. For high-content samples, the opposite is true and lower results are obtained.

In addition, different sample weighing directly affects the high-frequency induction combustion. The combustion temperature and state of the carbon-sulfur meter are not only related to the designed power of the high-frequency furnace itself, but also related to the amount of magnetically conductive substances that are induced. The combustion temperature is not constant, and as the material melts and burns, the amount of induction gradually decreases. For example, for some substances that are difficult to release carbon and sulfur (such as silicon, iron, etc.), since the weight of the added co-solvent is relatively constant, the heat generated is constant. At this time, too much sample will lead to incomplete combustion of the sample. The carbon and sulfur release is incomplete.

For instruments from different manufacturers, their linear ranges and calibration curve characteristics are different, and the implementation methods of high-frequency control parts are different, so the influence of sample weighing on the accuracy and reliability of analysis results is also different. for a specific instrument. It is necessary to familiarize with its characteristics through a large number of experiments, and choose the appropriate sample size to avoid the influence as much as possible.

The stacking order of samples and co-solvents

The cosolvent not only has the effect of increasing the magnetically conductive substances in the sample and increasing the combustion temperature, but also has the effect of increasing the fluidity of the sample and diluting the sample. During the analysis process, the stacking order of samples and fluxes directly affects the combustion results and analysis stability. For example, iron-based samples are directly burned by high-frequency induction under oxygen, and the reaction is violent, and the splash is serious after melting, which is likely to cause damage to the quartz tube of the combustion chamber and pollution of the ceramic protective sleeve. It was replaced with tungsten particles and the sample was placed on the upper layer. It was found that the quartz tube in the combustion chamber was quickly contaminated, and a thick layer of iron melt was adhered to the ceramic protective sleeve, which was difficult to clean, which not only affected the combustion tube. The service life also hinders the supply of oxygen, which directly affects the stability of the analysis results. The sample is placed on the bottom layer, and the tungsten particles are placed on the top layer. After the analysis, the quartz tube in the combustion chamber is very clean, and there is no metal melt splash on the ceramic thermal protective sleeve, and the analysis result is stable.

It can be seen that the influence of the stacking order of samples and fluxes on the stability of sample analysis results cannot be ignored.

Crucible

The blank of the crucible has always been a hot spot for Carbon Sulfur Determination. Untreated crucibles, blanks ranging from 10-100 μg/g. The pretreatment process conditions are set properly, and the blank of the crucible can be reduced to less than 1μg/g. Experiments show that the length of pretreatment time and temperature are very important to obtain a stable ground blank of the crucible. For example, when the crucible was fired at 800°C for 4 hours and used to analyze the steel sample, the obtained results fluctuated greatly. The blank of the crucible can be reduced to less than 1μg/g after being baked at 1100℃ for 4h.

Therefore, the crucible must be pretreated before use, and the appropriate baking temperature and time should be controlled. Generally, it is recommended to bake in a muffle furnace at 1100 °C for 4 hours, so as to minimize the impact of the blank of the crucible on the stability of the analysis results.

Reagents

The reagents used directly affect the stability and even the accuracy of the analytical results. In the use of high-frequency infrared carbon and sulfur analyzers, the dryness and purity of the analytical gas and carrier gas is the guarantee of reducing the blank of the system and obtaining accurate and stable analysis results. The purification process of the carrier gas first passes through alkali asbestos and then through magnesium perchlorate. In the experiment, if the installation order of the reagents in the reagent tube of the purifier is reversed, and the gas passes through magnesium perchlorate first, then through alkali asbestos, the measurement results will be inaccurate. The relative standard deviation increased. This is because alkali asbestos has a water absorption effect, but the water absorption efficiency is not as strong as that of magnesium perchlorate. Therefore, while the alkali asbestos absorbs the carbon dioxide contained in the gas, it will leak a small amount of water vapor, which will be brought into the analysis system by the carrier gas.

When the reagent (magnesium perchlorate) used to dry the analytical gas after combustion has failed, a small peak will appear on the sulfur baseline immediately after the high-frequency furnace is turned off. This is because the water vapor brought into the furnace by the outside air is not completely absorbed by the drying reagent, and the wavelength position of the infrared absorption characteristic peak of water vapor is similar to the infrared absorption characteristic peak position of sulfur dioxide, which directly affects the analysis results of sulfur, resulting in the analysis of sulfur. The curve is tailing and the result is high.

Dust

The adsorption caused by the accumulation of Huicheng during the analysis process is also an important factor affecting the stability of the analysis results, especially when analyzing low-content samples. The experimental results show that with the increase of the number of analyses, the deviation caused by dust becomes larger and larger. After the dust in the dust filter was thoroughly cleaned, the results of multiple analyses were consistent with the results of the first analysis. Therefore, during the analysis, the dust accumulation in the dust filter should be cleaned up in time.

Temperature

The influence of temperature on the stability of analysis results is mainly reflected in three aspects. First of all, the temperature of the dust filter is affected. In the case of the same amount of dust, the higher the temperature, the worse the adsorption effect. Secondly, the basis of gas analysis is inseparable from the gas state equation. The temperature controlled by the infrared analysis system is different. Causes the change of the volume of the analysis gas, thus causing the time difference for a certain amount of the analysis gas to pass through the fixed-length flood cell at different temperatures; in addition, the difference in the temperature of the infrared analysis system constant temperature control will cause the emission intensity of the infrared light source. Differences in the output of the electrical detector, thereby affecting the stability of the analysis results. Generally, the instrument will be equipped with constant temperature control for these parts. During the analysis process, the effect of constant temperature control should be monitored at any time to avoid the influence of temperature.

In summary, there are many factors that affect the stability of the analysis results of the carbon-sulfur analyzer. In the analysis process, especially when analyzing low-content samples, the influence of these factors can be comprehensively considered, so as to make an accurate evaluation of the analysis results. Avoid affecting the analysis results and progress due to wrong estimation of the problem, and increase the maintenance cost of the instrument.