Seestrasse 14B l CH-5432 Neuenhof l Tel. +41 56 406 12 12 l email. firstname.lastname@example.org
SPECTRAFLOW POTASH APPLICATION
The SpectraFlow Analyzer can be used to measure potash in several locations of the production process. An extensive Feasibility Study was done regarding the possibility and accuracy of the SpectraFlow measurement of Potash.
This report summarizes the results of investigations, which have been done on 5 different Potash samples. The main target is to clarify whether SpectraFlow can be used to analyze the finished product.
The customer, who provided the samples for this test has three production lines for Potash. One Hot Leaching Line (HLP) and two Cold Leaching Lines (NCCP and XCCP). For the future of this report XCCP will be replaced by CCCP.
It is of great interest for the customer to be sure that the final product conforms to the quality standards of each product. For this purpose the use of an On Line Analyzer on each of the feeding lines into the final silo is an investment of interest for the customer. At present only every hour a sample is taken.
As the customer has no experience in the use of Near Infra-Red (NIR) for the chemical analysis of Potash this prefeasibility study was done.
The results of this study is, that SpectraFlow can be used to measure potash in various production steps online.
Five samples of the final product were handed over to SFA. Two samples of HLP, called HLP_fine and HLP_standard, two samples of NCCP, called NCCP_fine and NCCP_granular and one sample of CCCP, called CCCP_fine.
The investigation focused on those questions, which can be answered by the samples without a detailed chemical analysis
• Does Potash show a sufficient structure in the Near Infra-Red range to enable an on line analysis?
• What is effect of the bright white color of Potash?
• What is the effect of temperature of Potash?
Methode and Results
The spectra were acquired with the test unit at the SFA premises in Switzerland.
The samples were measured by the test unit in the laboratory.
Of each sample 3 sets of spectra were taken: one set at ambient temperature (25°C), one set at 60°C and one set at 120°C. The spectra were taken at these three temperatue in order to check which effect the temperature of the Potash might have on the spectra.
The temperature levels were achieved by putting the samples together with their temperature resistant pan into the heating furnace, which is in the SFA laboratory and normally used to dry the samples. Because the total mass of the samples was not extremely big they were kept only for 15 minutes at the temperature levels described above. Again it shall be emphasized that the target was not to get out any residual water but to have a base for investigating the effect of higher Potash temperatures on the NIR spectra. While the spectra were acquired the material was kept inside the pans in which it had been heated up in order to avoid their cooling down during the acquisition time
The Potash samples are in general rather fine material. However as for all minerals the shape of the individual particles varies randomly and as a consequence their reflection behaviour may vary. In order to take this effect into account each sample was manually stirred 4 times at each temperature level and after the stirring the individual spectra were taken.
Figure 3 shows an example of the 5 acquired spectra sets. A typical spectrum at the 3 temperature levels is shown
Figure 3: Absorbance Spectra of HLP sample at the three temperature levels
Figure 4 shows how the spectra vary among each other at ambient temperature
Figure 4: NIR Spectra Examples for the 5 Potash Samples
The conclusions from the acquired spectra are very positive. The footprint is significant and shows a comparable pattern for all samples (The important aspect is they are comparable and not equal). Despite a comparable pattern each sample shows differences in the structure when investigating details. The spectra of the same sample at different temperature levels show greater similarity among each other than spectra of different samples at the same temperature level
The size of the sample particles has a rather strong influence on the NIR Spectrum. When comparing NCCP_fine with NCCP_granular the difference can be seen directly even when the offset is eliminated.
NCCP_fine shows a negative absorbance spectrum at the high wave number side. This is due to the bright white colour of the sample ,whose effect is of course more pronounced in the high wavenumber range, where NIR is already very close to visible light. It occurs occasionally for very bright white material (e.g. chalk) that the technically ideal white SPECTRALON is not as bright and white as the sample. In such a case the absorbance becomes negative. This means this phenomenon is not restricted to potash.
While the spectral structure of the individual samples maintain their characteristic shape when the temperature rises the “raw spectra” show offsets at the various levels
The global conclusion of acquiring spectra for Potash:
NIR CAN BE USED FOR ON LINE ANALYSIS OF POTASH
The present investigation confirms that the NIR spectra of finished Potash permit the use of SpectraFlow for the on line analysis of finished Potash.
In case of a definite commercial application two things need to be incorporated into the calibration
1. Training spectra of the same sample need to be acquired at the high temperature levels
2. At each temperature level the samples need to be mixed at least 4 times and after each mixing a spectrum needs to be acquired
The laboratory equipment of SFA permits to take all that into account.