XRF is one of the most commonly used technologies applying nuclear spectroscopy. It applies x-ray to excite electrons in the atoms of a sample, and the x-ray detector can determine the energies emitted from the excited atoms. This type of spectrum is called x-ray fluorescence spectrum, useful in quantitatively determining the identity and concentration of elements in an unknown sample. Common XRFs are either portable or as big as a desk.
The handheld XRF’s portability makes it useful in testing ores, metal pieces to be recycled, and other that need to be examined in mobile locations. However, handheld XRF has the disadvantage that it can only test a relatively small number of elements—normally it can only test metal products. In comparison, bigger table-size XRFs can have x-ray generators with higher power and other sophisticated sample preparation or testing process, allowing them to also test semi-metals or even non-metals in materials like plastic.
The first XRF device I got came from Niton (part of Thermo), model XLp818. Instead of an x-ray source, This device applies a 30mci americium radioactive source. The alpha ray can also excite the atoms and gain similar feedbacks as the characteristic energies are constant. This type of radioactive source containing device has been obsoleted now, but were common in the past. Compared to the x-ray tubes, the Am241 source has an advantage that it does not need to be replaced periodically as its half-life is more than 400 years. However, Am241 has its own characteristic x-rays, so it might obscure the performance when determining the peaks of some other elements where the sample’s peaks are covered by the source’s peaks.
This XLp originally cannot power on, a common problem due to the low storage time. This is normally caused by the unstable voltage due to temperature variances or battery problems. Moreover, it required me to replace the control board of the detector in the head of the device as the preamplifier voltage is higher than the rated voltage.
One important thing that should be noticed is that one should check the inner side of the plastic case under the screen. There is a glass tube containing red paint, and if its breaking indicates that the device has been dropped hardly. There is a severe risk of breaking the detector when the device is dropped. Referring to the silicon x-ray detector article, you will find that the crystalline detector is somewhat fixed on an electrical cooler. From experts repairing these XRFs, I learn that the crystal can sometime detach from the cooler and cause the detector to be useless when the machine is hit or dropped on the ground.
The second device I dismantled is SEA1000S, weighted more than 100Kg due to the thick shielding for the x-ray.
The cap can be opened to place the sample above the transparent window. There is a krypton membrane below to window to prevent damages to the detector and x-ray tube.
To enable the x-ray, the cap must be closed while a key is put in the hole at the right corner.
Open the case at the two sides of the machine, the main control board of the whole machine and the signal processing board for the x-ray detector can be accessible.
Power Supply & Signal Processor for the x-ray detector:
Main control:
In the center, the large gun-like item with a thick wire is the x-ray tube, model Varian VF-50J (maximum 50kV). The white thing is the thermal grease. There is also a fan right under it for the cooling of the tube.
The power supply comes from Spellman, one of the biggest HV generator supplier in the world. Model X3768, 4-50kV ~ 1mA.
The beryllium window of the x-ray tube will not directly touch the sample chamber. The x-ray is filtered by two modules—one with a tiny hole and the other with a yellow membrane. The first module is used to control the direction and area of the x-ray beam, and the second is used for protection I guess (but it is hard, unlike the protective krypton window).
The detector and the x-ray window is at a suitable angle to maximize the efficiency and minimize the escaping radiation, positioned by the copper module.
The detector, model XR-100CR from AmpTek, applies a Si-PIN detector. For the older detectors like this one, the case is made up of pure nickel (I used XRF to determine this) instead of stainless steel. I think that this is the reason why every detector from this type of machines I has a corroded surface.
A very good design is that a little cap made up of tungsten is applied on the head of the detector. This not only helps to protect the beryllium window, but also prevents the influence of background radiation or the x-ray escaped from the x-ray tube.
The detector is mounted on an aluminum case. The fan besides also helps it cool down.
Power supplies—12V, 24V.
Power supply division board.
Motor driver. A motor is used to control whether the x-ray window is hided. When there is no key or when the cap is not closed, a metal piece is moved by the motor to block the x-ray in case the x-ray is accidentally on.
With some modifications, I use a 160uci Am241 source from an ion mobility spectrometer as the x-ray source and the XR-100CR detector (the one in the picture is a newer one) to build a simple XRF system. The key is to make sure no radiation from the Am241 is received by the detector and affects the result.
Copper plate in the picture:
Stainless steel:
Every element has two characteristic x-ray peaks—Ka and Kb. The ratio between this two peaks can be used to determine the concentration of the element in a sample.
