Energy-Dispersive X-ray Fluorescence Spectrometer

(Stored in 0052 Arms)

General Summary

An x-ray fluorescence spectrometer is typically used for quantitative analysis of small volumes of solid and liquid samples, with a sensitivity that is superior to the electron microprobe, but inferior to an ICP-MS system. XRF analysis is non-destructive, to the extent that raw materials may be analyzed qualitatively without any sample preparation, although quantitative analysis of a rock powder may require a glass fusion to be made. A typical instrument description follows. The x-ray fluorescence spectrometer consists of a primary x-ray tube that emits characteristic x-rays of known energy. This primary x-ray beam is directed onto the surface of the sample; any elements in the sample having excitation energy below that of the primary beam energy will be fluoresced by the primary beam (and the most efficient excitation occurs for elements having excitation energy just below that of the primary beam energy). The fluoresced x-rays are then detected and counted by the x-ray detector, which is either a wavelength-dispersive spectrometer on a WDS-XRF system, or an energy-dispersive spectrometer on an EDS-XRF system. An EDS-XRF system typically makes use of a secondary target carousel that enables one to select a secondary target to be used for excitation of the sample, which allows for more efficient excitation of elements having a wide range of x-ray energies. The use of an energy-dispersive detector enables the simultaneous acquisition of x-rays over this wide energy range, and is most useful for quick screening of samples for a set of elements. Spectral acquisition times can be as short as a few seconds to the better part of 1000 seconds depending on the concentration of the elements being measured.

For a full discussion of the EDS-XRF technique, see the references for X-ray Fluorescence Spectrometry

Instrument Summary

The Kevex/IXRF 0700 x-ray fluorescence spectrometer is equipped with a Rh tube for primary excitation, and has a secondary target carousel containing 6 targets, resulting in a total of 7 analytical x-ray lines that can be used for excitation. The secondary target is irradiated by the Rh x-rays and emits the characteristic x-ray line of the target. Here is a list of the available excitation lines and their energies (the average of Ka1 and Ka2) and a list of elements that would be analyzed using them:

Target	Elements Excited by Target
Gadolinium Gd (42.761 Kev)	K-alpha lines of Cd In Sn Sb Te I Cs Ba La Ce (Pr)
Antimony Sb (26.274 Kev)	K-alpha lines of Ru Rh Pd Ag
Silver Ag (22.104 Kev)	K-alpha lines of Ga Ge As Se Br Kr Rb Sr Y Zr Nb Mo L-alpha lines of Ta W Re Os Ir Pt Au Hg Tl Pb Bi Th U
Germanium Ge (9.876 Kev)	K-alpha lines of Mn Fe Co Ni Cu Zn L-alpha lines of Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf
Iron Fe (6.400 Kev)	K-alpha lines of Sc Ti V Cr L-alpha lines of I Xe Cs Ba La Ce Pr Nd
Titanium Ti (4.508 Kev)	K-alpha lines of Cl Ar K Ca L-alpha lines of Ru Rh Pd Ag Cd In Sn Sb Te M-alpha lines of Bi Th U
Rhodium Direct Rh (2.696 Kev)	K-alpha lines of Mg Al Si P S Cl L-alpha lines of Ge As Se Br Kr Rb Sr Y Zr Nb Mo M-alpha lines of Er Ta W Re Os Ir Pt Au Hg Tl Pb

From this table one can see that the elements nominally from Magnesium through Uranium can be fluoresced using the appropriate secondary target. An energy-dispersive x-ray detector with a Beryllium window is used to detect x-rays emitted from the sample. This Be window absorbs low energy x-rays, and the low fluorescent yield of low atomic number elements (Na, Mg, Al, Si, for example) makes their detection problematic. The analytical sensitivity is best for elements with relatively high x-ray energy and minimal peak overlap, and poorest for elements with low x-ray energy and substantial peak overlap. In general, one should analyze the highest energy line for a given element to avoid overlaps and obtain the best count rate (i.e. K-alpha lines instead of L-alpha lines to avoid the overlapped energy range at low KV and to make use of the higher fluorescent yield of K-alpha lines).

The 0700 is outfitted with a roughing pump vacuum system, which should be used for measurements of lower energy x-rays. The sample must be fully contained in an appropriate sample holder to be used under vacuum.

The Rh x-ray tube can be used at accelerating voltages up to 60 KV and current values of up to 2.5 mA.

Software Summary

The Kevex/IXRF 0700 EDS-XRF x-ray spectrometer is controlled and run using the IXRF software, which allows one to perform completely automated spectral acquisition and quantification using multiple excitation targets and analysis conditions. The automation of these procedures is carried out using an analysis template, which contains the list of elements to be analyzed and the appropriate analysis conditions, background subtraction method, peak integration method, and quantification algorithm. Here is an example of an analysis template:

IXRF Analysis Template

The analyzed components are entered at the top of the form, and may include elements that are fixed in composition, or to be calculated by difference. The middle of the template contains the analysis condition to be used for that particular element, the peak intensity calculation method, and the quantification algorithm to be used to convert from x-ray intensity to concentration units. The bottom of the template has the list of analysis conditions and count times to be used during automated data collection.

Acquisition of an XRF spectrum is simple. Here are two screen captures illustrating the appearance of a spectrum. Spectra were acquired on two NBS glass standards, the first containing a nominal 500 ppm of about 60 elements in a Na-Al-Si-Ca matrix, and the second containing 50 ppm of these same elements. The analysis conditions used were Gd secondary target at 60 KV, sample under vacuum, with a count time of 250 sec (note that the strong Ag peak is due to both Ag in the standard and Ag solder in the target assembly containing the Gd secondary target). The peaks are quickly identifiable during acquisition for the 500 ppm standard, but it is evident from the spectrum for the 50 ppm standard that this is close to the detection limit for a number of elements. This illustrates that the general detection limit is several tens of ppm, depending on the nature of the sample (presence of overlapping x-ray lines and the nature of x-ray absorption or fluorescence from matrix elements are two important considerations).

NBS 500 ppm Glass Standard

NBS 50 ppm Glass Standard

The EDSXRF technique has very good sensitivity for (typically) high-energy x-ray lines that are not overlapped significantly, but relatively poor sensitivity for low energy x-ray lines. Elements having excitation energies just below that of the secondary target are generated most efficiently and can be detected at low concentrations. Those elements with lower excitation energies are not excited so efficiently, and require longer counting or else will not be detected at equivalent low concentrations. Notice the decreased intensity observed for Rb and Sr compared to Mo in the NBS 500 ppm standard, and the efficient fluorescence of Cs, Ba, and La immediately below the large Gd peak near 40 KV.

For optimum element suites the detection limit is as low as the ppm level, but in general the instrument shows good sensitivity at tens of ppm and higher. Peak overlaps and matrix effects can substantially raise the detection limit for an element. Na is not efficiently fluoresced, and results for Mg, Al, and Si may be better done using other techniques.

Quantitative analysis requires well-characterized standards of similar composition to the sample, and if a suite of standards can be acquired that bracketts the range of concentrations in the sample, good calibration curves can be obtained. A suite of USGS, NBS, and synthetic standards are available for use as x-ray fluorescence calibration standards. A number of the rock standards have been prepared as glass fusion discs, and rock powder standards are also available.

The IXRF software has algorithms for least squares analysis via a calibration curve assembled from measurement on standards, a fundamental parameters algorithm, and a matching algorithm that allows comparison with related spectra. See the IXRF manual for more information.

The EDS-XRF is typically used for screening materials to verify element profiles for further analysis. Minerals (both single crystal and powders), glasses (fragments and powders), gemstones, and liquids have all been analyzed on our system.

Sample Requirements

There are two large sample holders for solid samples, a 1.5 inch ring supported on very thin mylar x-ray film, and 1 inch capped cylinders with a lid that can be used for powder samples, also using the x-ray film for support. Solid samples may be placed under vacuum for analysis; powder samples can be placed under vacuum only if secured in these capped holders. Liquid samples cannot be placed under vacuum.

For qualitative XRF work a very small sample size may be used (down to about 1 mm), but the observed intensity is a function of the sample cross-section area presented to the x-ray beam. The larger the sample size, the larger the observed count rate.

Rock powders should be flux-melted to form glass discs for major element analysis. This reduces matrix effects and prevents problems due to grain-size sampling by the x-ray beam. Trace element analysis is usually carried out using pressed pellets of rock powder, in a matrix of cellulose and a briquetting material to hold the powder together.

XRF Use Policy

Currently the Kevex/IXRF is available on a first-come first-served basis, but the instrument can be reserved if necessary. All users must be certified as a Radiation Safety Worker in order to use the instrument. All users must have received an equipment checkout from Chi Ma in order to use the instrument.

Checkout Procedure

The Kevex/IXRF spectrometer is easy to use, especially with our new PC computer system. You can obtain an instrument checkout as follows:

First obtain a Radiation Safety Worker classification from the Safety Office (located in 25 Keith Spalding Bldg., contact Haick Issaian, Radiation Safety Officer, at x6727) in order to use the diffractometer unsupervised. This entails about 30 minutes and will require that you achieve a level of understanding of radiation safety issues; you will take a quiz administered by Safety in order to become a Radiation Safety Worker. If you already have Radiation Safety Worker status, this needs to be documented for the safety records kept in the Division Analytical Facility. Either have a copy of your paperwork made, or verify that your paperwork is on file in the Safety office.
Obtain a copy of the IXRF software manual. Currently you need to xerox the manual, but soon there will be an on-line manual that you can download. You need to read the manual to prepare yourself for the checkout so that you know what is being done.
Schedule time with the manager to get a checkout on the XRF spectrometer. This will require about an hour for discussion of safety issues related to the Kevex/IXRF 0700 unit, sample preparation, and use of the software.
The manager will place your name on the list of authorized users if and when you are able to correctly use the instrument and the software in an unsupervised capacity.

IXRF Manual

There is a manual which covers the operation of the IXRF software. You can make a xerox copy from a master for the time being. At some point a manual will be available from this web page.

Writeup Summary

Here is a summary of the instrument for the purposes of writeup.

Analyses are obtained from a Kevex 0700 energy-dispersive x-ray spectrometer, interfaced to an IXRF systems multichannel analyzer board and IXRF software. Primary excitation is supplied by a 2 watt Rh anode x-ray tube, and secondary excitation is obtained from a target wheel containing the secondary targets Gd (also contains Ag), Sb, Ag, Ge, Fe, and Ti. X-rays are detected using a Kevex energy-dispersive SiLi detector equipped with a Be window, and a Kevex 4405 pulse processor. A choice of peak area measurement and x-ray correction algorithms is available. Peak areas are determined using region-of-interest integration or gaussian deconvolution with background fitting. Spectra can be compared to other spectra obtained from standard materials via the match software. Quantitation of peak areas can be performed using the least squares calibration curve method, matching, or fundamental parameters algorithms.

Return to Division Analytical Facility