Introduction:
Currently, the materials generated during the zinc refinement process are analyzed with wet chemistry and LA-ICP-MS for CaO, MgO, SiO2, Al2O3, TiO2, Cr205 and F to monitor the refinement process. Typical concentrations are 0.5 to 35%. Four samples were sent representing various stages in the refinement process – Calcine, Zinc Ferrite Cake, Cobalt Cake and Copper/Cadmium Cake. The elements of interest for Calcine and Zinc Ferrite were Fe, Pb, S and Zn with the presence of Cu, Co, Cd, Ni, Pb and Zn of interest in the Cobalt and Copper/Cadimium Cakes.

Hardware Used:
LIBS2000+ broadband, high-resolution spectrometer
200 mJ Nd:YAG Big Sky laser
LIBS-SC sampling chamber with imaging module

Acquisition Parameters:
Laser setting 8 (highest setting ~200 mJ)
Analysis in air
-1 Q-switch setting for all samples except Calcine (Q-switch delay was increased to -2.5 setting to elminate the bright continuum background)
Element ID Parameters: +/-1 pixel search width and 50 count peak height

Measurement Mode:
LIBS

Experimental Conditions:
The four samples varied in color and consistency. Calcine was a brown powder, Zinc Ferrite was large chunks of brown material similar in consistency to clay and the Cobalt and Copper Cadmium Cakes were dark grey material with the consistency of thick mud. All four samples were analyzed on double-sided photo mounting tape adhered to glass microscope slides. Calcine and the Cobalt and Copper Cadmium Cakes were easily dispersed on the tape.

Due to an inability to spread the Zinc Ferrite sample, it was analyzed as large chunks. Three single shot spectra were acquired for different locations on each sample with two additional locations analyzed with 5 and 10 shot spectra for all samples except Calcine. Three and five shot spectra were used for Calcine due to the thin layer of sample present on the tape.

The use of 10 and possibly even 5 shot spectra would have drilled through the thin sample and into the tape and microscope slide. The other samples provided a thick enough layer that analysis of the underlying tape and microscope slide substrates was not a concern.

Results:
The results from the Elemental Identification software are summarized for each of the samples below. The complete list of elements identified by the software are found in an Excel spreadsheet with the results for each sample found on a separate worksheet.

Calcine
All the elements of interest (Fe, Pb, S and Zn) were detected in all replicates.
Decreased intensity and inability to detect S was observed when the dusty powdered sample coated the optics following the laser events. Compressed air was used to clean the sampling optics probe between each measurement.
Signal averaging did not improve detection (no additional lines were detected when spectra were acquired as the accumulation of multiple laser shots).

Zn Ferrite
Fe, Pb and Zn were detected in all replicates.
S was detected in one single shot replicate.
The sample was moist as observed visually and by the strong H alpha line at ~656 nm. If S was expected in this sample, the moisture in sample may have quenched the plasma making it more difficult to see some elements. Note that the Cobalt and Copper/Cadium Cakes appeared to have a higher moisture content.
Signal averaging did not improve detection.

Copper/Cadmium Cake
Less elemental lines were detected for the Copper/Cadmium Cake than for the Calcine and Zn Ferrite samples.
All elements of interest were detected in all replicates (Cd, Co, Cu, Ni, Pb and Zn).
The Cu lines were very intense (even stronger than the hydrogen alpha line).
The sample was moist as observed visually and by the strong H alpha line at ~656 nm.
Signal averaging did not improve detection.

Cobalt Cake
The least number of elemental lines was observed for the Cobalt Cake sample.
Cd, Co, Cu, Ni and Zn were detected in all replicates.
Pb was not detected in any of the replicates.
The sample was moist as observed visually and by the strong H alpha line at ~656 nm.
Signal averaging did not improve detection.