Inclusion Analysis of Metals Using Single Spark Evaluation Technology

The SPECTROLAB analyzer sets a new performance benchmark in spark emission spectroscopy. It provides the versatility of full spectrum coverage using CCD multichannel sensors in combination with photomultiplier tubes, allowing single spark evaluation (SSE) and time resolved spectroscopy.

Inclusions have a major impact on the mechanical properties of steel, aluminum and other metals. This article focuses on inclusion analysis using the SPECTROLAB analyzer. Firstly, we will outline the background of inclusion investigation using SSE technology. Then, the corresponding software routines will be explained.

As the first real hybrid system in the market, the SPECTROLAB uses the latest technologies that lead to enhanced performance in terms of precision, reproducibility and short and long term stability for metal analysis. In addition, the instrument offers lowest detection limits and reduced analysis times using the sophisticated read-out system coupled with the new plasma generator.

Besides achieving the lowest detection limits, Spark Analysis for Traces (SAFT) is utilized to reduce interelement effects. The variable plasma observation and newer spectral line selections provide higher precision of the measurements. SSE technology enables rapid detection of harmful inclusion populations in the sample.

Inclusion Analysis

Following the introduction of direct reading spark emission spectrometers, it was found that at the beginning of a spark period the intensities of certain elements displayed a significant super-elevation. After a few seconds, the intensities stabilized at a lower level.

Elements which demonstrated this effect strongly were calcium, aluminum, titanium, manganese and sulphur in steels and magnesium in cast iron. Later, advancements in electronics and software made it possible to record the intensities of every single spark.

For “normal” elements, the intensity development had a shape like the one shown in Figure 1. Following a small number of sparks, the intensity stabilized and remained at a constant level.

Single spark intensities for normal (dissolved) elements.

Figure 1. Single spark intensities for normal (dissolved) elements.

Example of single spark intensities for undissolved elements.

Figure 2. Example of single spark intensities for undissolved elements.

The reason behind different behaviors of these elements was easy to determine: The elements which demonstrated the initial super-elevation in intensity were those that form single element or single compound phases undissolved in the metallic matrix, as shown in Figure 2.

For instance, when an Al2O3 inclusion is hit by a spark, it will result in a high aluminum reading for this spark, because the aluminum concentration within the Al2O3 phase is roughly 53%.

However, the reason for the build-up of high intensities at the beginning of the spark process is still unknown. In order to answer this question, the impact of the individual sparks was analyzed.

Diameter and depth of a single spark crater.

Figure 3. Diameter and depth of a single spark crater.

The shape of a crater caused by a single spark of high energy is shown in Figure 3. The following observations were made:

  • The trace of the single spark is a crater comprising of a hole bounded by an elevated wall. It is logical to imagine that the wall is a zone, where the spark energy was adequate to melt the metal, but not sufficiently strong to vaporize it.

  • The entire structure has a diameter of about 34µm. The diameter of the hole, which matches with the zone of evaporation, is about 20µm. The hole’s depth is on the magnitude of 12µm.

Now it is easy to assume why the high readings are often found at the beginning of the spark process: If an inclusion was hit by a spark, it vaporizes and the inclusion is removed from the surface of the sample. This holds true as long as the spark crater is bigger than the inclusion.

When there are many inclusions of a specific composition, larger than the crater hole, information regarding their number and size cannot be obtained from the spark signals. The dimensions of the spark crater differ with respect to the spark’s energy; the higher the single spark energy, the larger the spark crater.

Statistics

Let us imagine that the spark craters are considerably larger than the inclusion under observation. By observing the intensities of spark number 100 and following In Figure 1, we can see that they have an average intensity of about 1770. In statistics, it is easier to present the data as a histogram.

Histogram of the intensities of a dissolved element.

Figure 4. Histogram of the intensities of a dissolved element.

A histogram with class widths of 9 digits is shown in Figure 4. 35 classes are displayed. The histogram is based on a dissolved element’s intensities. This means, the same type of element delivered the information for Figure 1.

There is a specific reason why extreme care was taken in describing the distribution of the intensities of an undissolved element: Majority of elements that form inclusions are partly dissolved in the metallic matrix. The metallurgical effects of the inclusion portion and the dissolved elements are entirely different. Hence, it is useful to differentiate between the two phases.

Determination of Particle Counting and Particle Sizes

As stated above, inclusions have a major influence on the mechanical properties of metals such as steel, aluminum, etc. Here, the inclusions of manganese sulfide lead to short turnings. In applications where inclusions play a critical role, it is essential to control homogeneity, shape and size of the inclusions. Often, the most vital criterion is the absence of any inclusions.

Table 1. Weighting factors for inclusions of size class 0-8.

0 1 2 3 4 5 6 7 8
0.05
0.1
0.2
0.5
1
2
5
10
20

For each size class, the inclusion count is multiplied by a factor, as shown in Table 1, and the results are added. It is a good indicator of the sample’s cleanliness.

Table 2. Inclusion sizes of different sizes and classes.

Class Inclusion length Inclusion diameter
Sulfidic Oxidic inclusions
Type SS Type OA Type OS Type OG
0 1 2 3 4 5 6 7 8 9
0 60-90 40-60 60-90 40-60 27-40 40-60 27-40 18-27 9-13 5-7
1 90-140 60-90 90-140 60-90 40-60 60-90 40-60 27-40 13-19 7-9
2 140-210 90-140 140-210 90-140 60-90 90-140 60-90 40-60 19-27 9-13
3 210-310 140-210 210-310 140-210 90-140 140-210 90-140 60-90 27-38 13-19
4 310-470 210-310 310-470 210-310 140-210 210-310 140-210 90-140 38-53 19-27
5 470-700 310-470 470-700 310-470 210-310 310-470 210-310 140-210 53-76 27-38

In Table 2, only the rows 0 to 5 are listed; rows 6 to 8 cover inclusions that are too large for a single spark evaluation.

After knowing the span of inclusions that can occur, analysis of single spark can help determine the cleanliness of steels. If a large inclusion, larger than the spark crater, is hit by a spark the same maximum intensity is expected.

However, the spark tends to hit the border between metallic phase and inclusion. Inclusions that surpass a certain size limit always offer the same intensity readings, irrespective of how large they are.

Ablated portion of a big and a very big inclusion.

Figure 5. Ablated portion of a big and a very big inclusion.

If an inclusion, considerably smaller than the spark crater, is hit the signal tends to become weaker. The signal can be utilized to determine the inclusion’s size (Figure 5).

SPECTROLAB Technology

The instrumental requirements for this approach are accurately controlled and reproducible spark conditions, high optical resolution and the ability to utilize closely adjacent emission lines; excellent performance in the ultra-violet region of the emission spectrum; and sophisticated results processing.

Schematics of the SPECTROLAB analyzer.

Figure 6. Schematics of the SPECTROLAB analyzer.

SPECTROLAB employs a version of the SPECTRO Spark Analyzer Vision software for Windows. This comprises an advanced calibration module, an integrated SQL database and easy-to-use operator interface. Preset analytical methods can be used to configure the inst¬rument, so that for regular analysis, difficult measurement and calculation procedures can be performed easily.

Conclusion

There is one application where the diameter or size of the inclusion can be determined: In clean steels without a major fraction of inclusions greater than class 3, the sizes and counts of the OG-Type inclusions can be determined. This is not possible for other inclusion types and sizes classes.

However, it is possible to estimate a value characterizing the cleanliness of the steel, akin to that of a K0 test according to DIN 50602 standards. This is performed by forming single spark histograms and subtracting the Gaussian distribution of the metallic phase from the total. Both options are supported by the SPECTROLAB software.

This information has been sourced, reviewed and adapted from materials provided by SPECTRO Analytical Instruments GmbH.

For more information on this source, please visit SPECTRO Analytical Instruments GmbH.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    SPECTRO Analytical Instruments GmbH. (2019, August 26). Inclusion Analysis of Metals Using Single Spark Evaluation Technology. AZoM. Retrieved on November 13, 2019 from https://www.azom.com/article.aspx?ArticleID=11149.

  • MLA

    SPECTRO Analytical Instruments GmbH. "Inclusion Analysis of Metals Using Single Spark Evaluation Technology". AZoM. 13 November 2019. <https://www.azom.com/article.aspx?ArticleID=11149>.

  • Chicago

    SPECTRO Analytical Instruments GmbH. "Inclusion Analysis of Metals Using Single Spark Evaluation Technology". AZoM. https://www.azom.com/article.aspx?ArticleID=11149. (accessed November 13, 2019).

  • Harvard

    SPECTRO Analytical Instruments GmbH. 2019. Inclusion Analysis of Metals Using Single Spark Evaluation Technology. AZoM, viewed 13 November 2019, https://www.azom.com/article.aspx?ArticleID=11149.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback
Submit