Skip to end of metadata
Go to start of metadata

X-Ray Fluorescence

Introduction to X-Ray Fluorescence

XRF is an elemental analysis technique based on the emission of characteristic "secondary" X-rays from a material. It is a non-destructive way of doing chemical analyses of rocks, minerals, sediments and fluids. The wide applicability of this method is due to low cost sample preparation and ease of use. It can trace the major elements of rocks and metals with an accuracy of ~ 0.1%.

Operating principle

When materials are bombarded with high energy, short wavelength radiation such as X-ray, they become ionized by losing the inner shell electrons. This bombardment can be seen in the first two steps of Figure 1. An outer shell electron fills the vacancy that has been created from the removal of the inner shell electron as shown in the final step. By doing this process there is energy released in the form of another x-ray (this is the fluorescence) which is equal to the difference in energy between the two shells. The wavelength of this fluorescence emitted is characteristic and unique to each element. The intensity of the detected signal is proportional to the concentration of the elements.


Figure 1. Showing 3 step process for XRF. a) Beam hits electron b) Electron is kicked out of shell c) Another electron drops producing a fluorescence in the process.

Example XRF instrumentation

The unit X-met 3000TXV+ is a portable X-Ray Fluorescence (XRF) analyzing tool. It includes a silver x-ray tube and an analyzer in a single package that is also portable. Launched by Oxford Instruments for the analysis of light metal alloys. This allows the measurement of all important elements in the aluminum and titanium alloys. Traditional portable XRF analyzer can not detect all the elements in aluminum alloys because magnesium, aluminum, silicon produced by low energy X-ray. X-MET3000TXV + incorporated a vacuum pump to achieve the detection of important elements to improve the performance of the analyzer

X-Ray Fluorescence Project


Project team. Left to Right. Chandan Roy, Damian Marrufo

Introduction

X-ray fluorescence has been used to do quantitative chemical analysis of artifacts avoiding destructive testing. In this case ceramic samples from pottery of regional tribes have been gathered and will be analyzed to try to discover correlation between them. This information will allow regional historians to be able to trace movements of people and possible trading routes among other information.

Material and Methods

Three ceramic samples were provided for this study. The three samples were labeled 17297E 17153B and 17153R they were further identified as unknown redware for the first two and Three Rivers for the last. A picture of the samples is provided in Figure 1.


Figure 1. Pictures of three ceramic samples that were analyzed.

The x-ray tube source and XRF analyzer was all provided by a hand-held portable unit X-met 3000TX which creates x-ray by way of a Ag target. The device was run at a setting of 40 kV and 7mA and was allowed to collect data during 300 seconds for each of the three samples. The data was then processed using PyMca which is a program to process XRF data.

Results and Discussion

The collected data was interpreted using PyMca. The data was first calibrated choosing the final high energy peak as silver Ag KL2. Afterwards a fit was performed specially to identify the largest peaks inside the sample and not the overall composition. The plots for the three samples are shown in Figure 2.




Figure 2. Data plots for 17297E, 17153B and 17153R samples respectively in log scale showing source data in black, fitted curve in red and background in yellow.

After a good fit was achieved the final report on the peak data and mass fraction was obtained. The information is shown in the following tables for each of the 3 samples.

Table 1. Peak data and molar concentration for 17297E

Element

Group

Fit_Area

Sigma_Area

Mass_fraction

Fe

K

1.72E+005

5.61E+002

0.0009572000

Sr

K

3.06E+004

2.28E+002

0.0000142600

Zr

K

2.06E+004

2.17E+002

0.0000076650

Ag

K

1.36E+004

2.16E+002

0.0000028710

Au

L

8.87E+003

1.40E+002

0.0000085920

Rb

K

4.96E+003

1.36E+002

0.0000026090

Nb

K

3.63E+003

1.48E+002

0.0000012190

Ti

K

3.55E+003

6.72E+001

0.0002213000

Y

K

3.45E+003

1.37E+002

0.0000014240

Ca

K

3.14E+003

6.70E+001

0.0016130000

K

K

2.96E+003

6.32E+001

0.0074480000

Zn

K

2.31E+003

5.98E+001

0.0000037660

Mn

K

7.94E+002

8.01E+001

0.0000070880

Cr

K

5.83E+002

4.00E+001

0.0000087410

As

K

5.38E+002

5.38E+001

0.0000004977

Cu

K

5.25E+002

3.22E+001

0.0000011070

W

L

4.85E+002

6.83E+001

0.0000007460

Ni

K

2.42E+002

2.56E+001

0.0000006495

Ga

K

2.18E+002

3.47E+001

0.0000002927

Table 2. Peak data and molar concentration for 17153B

Element

Group

Fit_Area

Sigma_Area

Mass_fraction

Fe

K

1.48E+005

4.06E+002

0.0008241000

Sr

K

2.67E+004

2.12E+002

0.0000124500

Zr

K

1.83E+004

2.06E+002

0.0000068160

Ag

K

1.18E+004

2.06E+002

0.0000024880

Au

L

8.62E+003

1.38E+002

0.0000083510

Rb

K

3.87E+003

1.29E+002

0.0000020380

Nb

K

3.57E+003

1.44E+002

0.0000011990

Y

K

3.39E+003

1.34E+002

0.0000013980

Ti

K

3.22E+003

6.60E+001

0.0002008000

Ca

K

2.91E+003

6.41E+001

0.0014960000

K

K

2.36E+003

5.74E+001

0.0059280000

Zn

K

2.34E+003

5.98E+001

0.0000038230

As

K

5.70E+002

5.28E+001

0.0000005275

Mn

K

4.63E+002

6.09E+001

0.0000041310

Cu

K

4.10E+002

3.03E+001

0.0000008649

Cr

K

3.94E+002

4.46E+001

0.0000059070

W

L

3.31E+002

7.21E+001

0.0000005089

Ga

K

2.08E+002

3.19E+001

0.0000002799

Ni

K

1.91E+002

2.50E+001

0.0000005133

Table 3. Peak data and molar concentration for 17153R

Element

Group

Fit_Area

Sigma_Area

Mass_fraction

Fe

K

1.23E+005

4.76E+002

0.0006865000

Sr

K

2.60E+004

2.16E+002

0.0000121200

Zr

K

2.39E+004

2.28E+002

0.0000088890

Ag

K

1.31E+004

2.20E+002

0.0000027580

Au

L

9.08E+003

1.44E+002

0.0000087930

Rb

K

5.39E+003

1.38E+002

0.0000028370

Ca

K

4.47E+003

7.70E+001

0.0022960000

Nb

K

4.17E+003

1.52E+002

0.0000014000

Y

K

3.53E+003

1.40E+002

0.0000014560

Zn

K

3.40E+003

7.06E+001

0.0000055520

K

K

3.40E+003

6.75E+001

0.0085600000

Ti

K

3.32E+003

6.47E+001

0.0002071000

Mn

K

1.20E+003

8.12E+001

0.0000106800

As

K

8.79E+002

5.75E+001

0.0000008136

Cr

K

5.24E+002

3.92E+001

0.0000078470

Cu

K

4.61E+002

3.16E+001

0.0000009730

W

L

3.34E+002

7.59E+001

0.0000005138

Ga

K

2.23E+002

3.62E+001

0.0000002992

Ni

K

2.03E+002

2.47E+001

0.0000005437

Notice the fact that even though Fe has the highest peak count (given by the fit area) it does mean the composition of the clay is mostly iron as shown in the mass fraction field. If the curve on the left of the plotted data in Figure 2 were taken into account, other elements(that make up a larger mass fraction of the sample) would also be included like oxygen and aluminum among others. Since this is not the objective and we simply want to find correlation between the highest peaks this information was ignored. The second to last curve to the right of plotted data in Figure 2 was also ignored since it does not match the shape of a XRF curve.

Now we will take the peaks with the highest count and make data plots between them for the three samples. All of these plots will be identified in group has Figure 3.


Figure 3. Data plots comparing number of counts for highest peaks

After all of these graphs we can see that even though the original data plots in Figure 2 looked very similar, when comparing the relative counts for each of the highest peaks we have greater variation that can lead us to make an educated conclusion

Conclusion

Comparing the data that was obtained through the XRF analyzer and looking at the peaks, it is possible to conclude that the clay samples 17153B and 17297E were produced from a very similar or even same source and following the same work method. The reason we can conclude this is by looking at all of the data plots shown in Figure 3. It can be seen that the relationship between all of the counts of 17153B and 17297E follow a linear upward relationship. That means that the detector was always reading similar ratios of x-rays from XRF from the two samples. So looking at all of the plots in Figure 3, the XRF analyzer unit detected less counts from all of the elements from 17153B then from 17297E.

Similar we can conclude that 17153R is completely foreign to the other two samples, the reason being that it did not follow a similar ratio. The data points on it would sometimes indicate an upward linear relationship while other times it demonstrates a downward relationship. A downward relationship would indicate the more counts you have of one element, the less counts you will have of another. This would indicate a different source and/or work method.

Additional information

Figures for configuration of fit

 

  • No labels