Raw materials XRF application for the cement industry using a universal borate fusion methodology

Introduction

A previous paper portrayed sample preparation by fusion methodology and the X-ray fluorescence (XRF) spectrometry analytical conditions for the calibration of cement materials^[1]. Experiments were conducted according to the guidelines of two well-known cement chemical analysis International Standard Methods and results were presented. The results proved that this robust analytical method is able to qualify by the ASTM C 114^[2] and ISO/DIS 29581-2^[3] norms.

This analytical method was developed using a Claisse® M4^TM fluxer, an automated fusion instrument for the sample preparation and a Wavelength Dispersive X-Ray Fluorescence (WDXRF) spectrometer, for the determination of all the elements of interest relating to the cement industry. The method was used to prepare finished products, process materials, as well as a very large range of raw materials which will be described in this second paper. Cement, blended cement, cement with additions, aluminate cement, clinker, kiln feed, raw mix, limestone, gypsum, sand, clay, bauxite, silica fume, slag, fly ash and iron ore are among the raw materials covered by this analytical application. This sample preparation was also transferred and optimized for TheOx® fluxer, a fully electric fusion instrument, by executing a robust validation of the method’s performance.

Due to the fact that powders are affected by mineralogy and particle size effects^{[4, 5]}, it is almost impossible to use a single XRF calibration curve for the analysis of such a wide range of different materials using the pressed powder preparation method. When fused in a borate glass, all mineralogy and particle size effects are eliminated. A single XRF calibration curve covering the whole range of concentrations for all elements of interests for the cement industry raw materials analysis can be made.

This paper examines all the XRF analytical conditions for the calibration of the entire range of raw materials using the robust borate fusion sample preparation methodology as well as the numerous Reference Materials (RMs) used for this analytical application. The results of this general and unique XRF raw materials calibration will also be presented in terms of precision, accuracy, and limit of detection.

Experimental

Apparatus and instrumental conditions 

All information regarding instruments, sample preparation methodology development, final optimized conditions of using a Claisse^® M4 fluxer and robustness analysis of the preparation method for sample preparation by fusion was presented in the previous paper^[1].

In the following, a sequential WDXRF spectrometer with a rhodium end-window X-ray tube of 1000 watts was used for data generation. The spectrometer analytical conditions, peak-line, background measurements, background position, pulse-height, counting time and other parameters were defined and optimized by the wavelength step-scanning of standard disks representative of the application. The spectrometer analytical conditions for the measurement of all the elements used for the raw materials application were presented in a previous paper^[6]. The analytical lines for certain elements were added to the analysis method because the reference values for these elements were available from the raw materials RMs. The measurements were executed under vacuum using a 28 mm collimator mask.

Calibration preparation

The calibration of the WDXRF raw materials application was executed using a wide variety of RMs from the following origins:

• Bureau of Analysed Samples Ltd. (BAS): British Chemical Standard Certified Reference Materials 
• Domtar® Inc. Research Center: Canadian Certified Reference Materials 
• China National Analysis Center for Iron and Steel: NCS DC Reference Material 
• European Committee for Iron and Steel Standardization 
• European Coal and Steel Community (ECSC): Euro-Standard 
• Geological Institute for Chemical Minerals: GBW Reference Material 
• Institut de Recherches de la Sidérurgie (IRSID) : Échantillon-Type
  
• Instituto de Pesquisas Tecnológicas (IPT) : Reference Material 
• Japan Cement Association (JCA): Reference Materials for Xray Fluorescence Analysis
• National Institute of Standards & Technology (NIST): Standard Reference Material® 
• Slovak Institute of Metrology (SMU): Slovak Reference Material 
• South Africa Bureau of Standards: SARM Certified Reference Material

Table 1 demonstrates the certified element concentration ranges in both the original sample base and the ignited base.

Table 1. RM element concentration as an oxide equivalent

Compound	Concentration Range of the Certified Reference Materials
	Original Sample Base  (%)	LOI Free Base (%)
SiO2	0,02 - 99,78	0,03 - 99,86
Al2O3	0,004 - 85,07	0,005 - 85,32
Fe2O3	0,005 - 85,3	0,005 - 91,03
CaO	0,006 - 70	0,006 - 98,58
MgO	0,001 - 21,25	0,001 - 39,66
SO3	0,02 - 46,3	0,02 - 58,54
Na2O	0,001 - 4,81	0,001 - 4,84
K2O	0,001 - 4,99	0,001 - 5,02
TiO2	0,004 - 3,76	0,004 - 3,77
P2O5	0,003 - 8,42	0,003 - 8,62
Mn2O3	0,0001 - 4,93	0,0002 - 5,05
SrO	0,001 - 0,638	0,001 - 0,649
Cr2O3	0,0002 - 0,474	0,0004 - 0,486
ZnO	0,0001 - 0,107	0,0001 - 0,109
ZrO2	0,005 - 0,14	0,005 - 0,2
V2O5	0,0006 - 0,72	0,0007 - 0,75
BaO	0,0012 - 0,66	0,0012 - 0,66

Two sets of the different standard glass disks were produced on both the M4TM and the TheOx® fluxers. The first set was used for calibration. Once the calibration was completed, the two sets of standard glass disks were analyzed as unknown. The results were then used to evaluate the precision and the accuracy of the methodology.

Results and discussion

Robustness of the fusion method

As previously published^[1], it was determined that ignition of the sample is absolutely necessary in the analytical process for a global fusion method. This critical step allows fusion of the raw materials and cements with additions, which are difficult or impossible to fuse in the non-ignited state with traditional fusion methods. A preparation with a sample to flux ratio of 1:10 requires a fusion program of approximately 13 minutes of heating at more or less 1050°C to prepare stable glass disks with high alumina and/or high silica samples. The cooling process is attained with forced air in 5 minutes. 

To demonstrate the robustness of this fusion methodology, more than 200 different samples from 20 different material types were fused with the global fusion method.
The materials are listed in Table 2. This list includes materials not commonly used as raw materials, but sometimes found in waste materials and used as fuel to test the limits of the global nature of the method.

Table 2. List of materials used in this experiment

#	Material Type	Tried	Success
1)	Cement	118	118
2)	Cement with AdditionsA	15	15
3)	Aluminate Cement	7	7
4)	Clinker	13	13
5)	Kiln Feed/Raw Mix	11	11
6)	Limestones	9	9
7)	Gypsum	7	7
8)	Clay	7	7
9)	Bauxite	4	4
10)	Sand	7	7
11)	Silica Fume	3	3
12)	Fly Ash	9	9
13)	SlagB	8	6
14)	Iron OreC	19	12
15)	Other	5	5
	Overall	242	233

^{A. Only the cements with known additions are listed here; the cement category probably included some cements with additions
B. The two slag samples that failed contained a higher level of copper
C. The iron ore samples that failed contained a higher level of copper}

The Global Fusion Method concluded good efficiency to prepare homogenous and stable lithium borate glass disks with all of the materials except for three (3): iron ores with high magnetite content, copper-rich iron ores and copper slags.

Calibration

Table 3 enumerates the inter-element corrections that were used as well as their type. Also displayed are the squared correlation coefficients from the calibration curves of the analyzed elements for the optimized selection of RMs.

Table 3. Inter-element corrections and squared correlation coefficients for raw materials application

Element Line	Inter-element Correction Information		Squared Correlation Coefficient
Al Kα	Fixed Alphas	---	1,0000
Ba Lα	Fixed Alphas	Overlap for Ti	0,9982
Ca Kα	Fixed Alphas	---	1,0000
Cr Kα	Fixed Alphas	Overlap for V	0,9997
Fe Kα	Variable Alphas	---	1,0000
K Kα	Fixed Alphas	---	0,9999
Mg Kα	Fixed Alphas	---	0,9999
Mn Kα	Fixed Alphas	Overlap for Cr	0,9998
Na Kα	Fixed Alphas	---	0,9998
P Kα	Fixed Alphas	---	1,0000
S Kα	Variable Alphas	---	1,0000
Si Kα	Variable Alphas	---	1,0000
Sr Kα	Fixed Alphas	---	0,9995
Ti Kα	Fixed Alphas	---	0,9999
V Kα	Fixed Alphas	Overlap for Ba & Ti	1,0000
Zn Kα	Fixed Alphas	---	0,9986
Zr Kα	Fixed Alphas	Overlap for Sr	0,9988

From the calibration of the different elements, three (3) correlation curves of corrected concentrations vs certificate concentrations are of particular interest. The CaO curve (Figure 2), since it is the major constituent of cement products, the SiO₂ curve (Figure 3), because silica variable mineralogy influences the XRF count intensities^[4], and the SO₃ curve (Figure 4), seeing that sulfur, under its different oxidation states, has the reputation to be a volatile compound^[5].
Figure 1. CaO correlation curves of corrected concentrations vs certificate concentrations

Figure 2. SiO₂ correlation curves of corrected concentrations vs certificate concentrations

Figure 3. SO³ correlation curves of corrected concentrations vs certificate concentrations 

Sensitivity, precision and accuracy results

Table 4 illustrates the results obtained from assessing sensitivity, precision and accuracy. Using the spectrometer software, the sensitivity results were captured using the Lower Limit of Detection (LLD). Precision was evaluated on an absolute concentration base (%) by calculating the maximum difference between the results of the analyzed elements for the duplicate preparations of all referencematerials used in the calibration. The precision evaluation was executed for both the M4TM and the TheOx® fluxers. The accuracy evaluation was determined on an absolute concentration base (%) by calculating the maximum difference of the two results obtained from the duplicates on both the M4TM and the TheOx® fluxers, against the certified value over all the reference materials used in the calibration. The standard deviation was calculated by the software and is also presented in this table.

Table 4: Raw materials application results

Compound	LLD  (ppm)	Max. Dev.A Between  Duplicates  M4  (%)	Max. Dev.A Between  Duplicates  TheOx (%)	Software  Standard Deviation (%)	Max. Dev.A  from  Certified  Value (%)
SiO2	40	0,10	0,15	0,16	0,59
Al2O3	61	0,11	0,16	0,105	0,34
Fe2O3	51	0,12	0,11	0,089	0,36
CaO	41	0,17	0,19	0,16	0,34
MgO	84	0,04	0,04	0,081	0,22
SO3	47	0,11	0,05	0,062	0,19
Na2O	85	0,02	0,02	0,018	0,07
K2O	17	0,02	0,02	0,011	0,04
TiO2	42	0,02	0,01	0,0130	0,06
P2O5	58	0,02	0,03	0,0074	0,03
Mn2O3	25	0,009	0,012	0,012	0,04
SrO	14	0,01	0,006	0,0044	0,014
Cr2O3	34	0,006	0,005	0,0024	0,008
ZnO	13	0,003	0,002	0,0010	0,003
ZrO2	11	0,002	0,002	0,0032	0,007
V2O5	15	0,003	0,001	0,0019	0,004
BaO	57	0,01	0,006	0,0060	0,014

^{A. Max. Dev. = Maximum Deviation.}

The results proved excellent accuracy and precision despite the wide range of elements.

Conclusions

A global fusion/XRF analytical method for cement industry materials has been described in this paper as well as in the previous paper^[1]. This method of preparation by fusion allows fusing cements and all raw materials that are normally found in a cement plant. The cement
calibration complies with the precision and accuracy requirements of the International Standard Methods for cement analysis (ISO/DIS 29581-2 and ASTM C 114).

In this paper, it was proven that it is possible to obtain a precise and accurate universal calibration, able to cover the wide range of raw materials used by the global cement industry. The range of materials include cement, blended cement, cements with additions, aluminate cement, clinker, kiln feed, raw mix, limestone, gypsum, sand, clay, bauxite, silica fume, slag, fly ash and iron ore, just to name a few.

References

1. BOUCHARD, M., ANZELMO, J.A., RIVARD, S., SEYFARTH, A., ARIAS, L., BEHRENS, K., DURALI-MÜLLER, S., “Global Cement and Raw Materials Fusion/XRF Analytical Solution”, Advances in X-ray Analysis, Vol. 53, Proceedings of the 58th Annual Conference on Applications of X-ray Analysis (Denver X-ray Conference), International Centre for Diffraction Data, ISSN 1097-0002, 2010, pp. 263-279.
2. ASTM, Standard C 114-08, “Standard Test Methods for Chemical Analysis of Hydraulic Cement”, Annual Book of ASTM Standards, Volume 04.01, ASTM International, West Conshohocken, PA, 2008, pp. 150–157.
3. DIN EN ISO 29581-2 (Draft standard, 2007-07), “Methods of testing cement - Chemical analysis of cement - Part 2: Analysis by X-ray fluorescence” (ISO/DIS 29581-2:2007), 30 pp.
4. ANZELMO, J.A., “The Role of XRF, Inter-Element Corrections, and Sample Preparation Effects in the 100-Year Evolution of ASTM Standard Test Method C114”, Journal of ASTM International, Vol. 6, No. 2, Paper ID JAI101730, available online at www.astm.org, 2009, pp. 1-10.
5. SPANGENBERG, J. and FONTBOTÉ, L., “X-Ray Fluorescence Analysis of Base Metal Sulphide and Iron-Manganese Oxide Ore Samples in Fused Glass Disc”, X-Ray Spectrometry, Vol. 23, 1994, pp. 83-90.
6. BOUCHARD, M., ANZELMO, J.A., RIVARD, S., SEYFARTH, A., ARIAS, L., BEHRENS, K., DURALI-MÜLLER, S., “Global Cement and Raw Materials Fusion/XRF Analytical Solution. II”, Powder Diffraction, Vol. 26, No. 2, International Centre for Diffraction Data ISSN 0885-7156, 2011, pp. 176-185.0,0002 - 5,05