Ultra-Low kV EDS
Figure 4 : Schematic diagram showing the difference in geometry between a windowless detector with a non-circular sensor and a circular large-area SDD detector. A conventional detector with a circular sensor operates at a longer working distance and further from the sample.
Figure 3 : X-ray spectra of Al metal with its native oxide taken at 2 kV using two windowless EDS detectors. The spectrum in yellow is with the new low-noise electronics and that in red is with standard electronics.
K-series of iron also decreases [ 2 ]. T is highlights two issues. Firstly, the K-series is only well excited at high kV (for example, 20 kV) and has low intensity at 10 kV. Secondly, below 7 kV only the less reliable L-series is excited, and its intensity also decreases rapidly as the accelerating voltage is reduced. At 2 kV the intensity of Fe L is around 35× smaller than Fe K intensity generated at 20 kV. Furthermore, a low accelerating voltage for high spatial resolution also means a small beam current, oſt en measured in 10s or 100s of pA, thus further reducing the X-ray count rate. T e largest conventional EDS detectors off er up to 15× the sensitivity of the 10 mm 2 detectors that were typically used on a tungsten fi lament SEM (W-SEM) before the introduction of the SDD. T is goes someway to off set this count-rate defi cit, however even the largest conventional detectors are only practical for analysis down to about 3 kV accelerating voltage, which corresponds to 30–50 nm spatial resolution ( Figure 1 ). Hardware developments for low-kV elemental analysis . Space constraints limit the practical size and distance from the specimen of even SDD sensors. Therefore to improve sensitivity, other methods have been investigated. One way to improve X-ray intensity for low-energy lines is to remove the X-ray window from the front of the detector. Although windowless SDD detectors have been available for TEM, removing the window presents design challenges for the SEM because of the potential for contamination of the sensor at poorer vacuum levels and the higher frequency of chamber venting. However, compared to liquid-nitrogen-cooled Si(Li) detectors, SDD detectors can operate at much higher tempera- tures, and the thermoelectric cooling employed off ers much
Improvement 22
Si L l ×8
faster sensor warming and cooling. T e detector therefore can be switched off automatically either by interlocking the pump/ vent cycle of the microscope or by sensing changes in vacuum at the detector. Provided the electronics for the detector are designed to achieve signal-measurement stability very rapidly after cool down, a windowless SDD can be warmed up for specimen exchange and be back in operation within minutes. Removing the window removes the absorbing eff ects of the window material and also the obscuring effect of the grid structure that supports a thin window. T e typical benefi ts of this for a windowless design versus a grid-supported polymer window can be seen in Table 1 . For microanalysis using K lines at high kV, count rates with a windowless detector can be up to 50% higher. For low kV EDS, count rate increases for lines less than 1 keV are greater, with typical improvements about 2–3×. For very-low X-ray energies (sub-100 eV), improvements rise rapidly, making practical the detectability of low-energy L-lines of Si, Al, and Mg, and even the K-line of Li. T e Si L and Al L lines are particularly important because their K lines are not usefully excited below 2 kV, and they are common constituents of many nano-materials.
Removing the window is not sufficient for successful low-energy EDS because the very low-energy photons still must be detected and measured. The efficiency of a conventional electronics counting chain (detector and pulse processor) is unlikely to be good enough for extremely low-energy X-rays. Low-energy noise has to be minimized and signal processing must be optimized to separate from noise the lowest energy X-rays such as Li K and Al L (Figure 3). One potential additional benefi t of removing the window is the saving of space, allowing the sensor to be moved closer to the sample. Thus, a further improvement in sensitivity can be achieved by using an oval-shaped sensor instead of a circular device. This allows a detector of the same large area to be positioned closer to the sample without touching a conical pole-piece ( Figure 4 ). Using an oval-shaped design in a windowless confi guration, solid angles over 0.2 sr have been achieved in practice on FEG-SEMs. Taking into account
Table 1 : Improvement in count rate for selected X-ray lines with a windowless detector design versus a conventional detector with an AP3-type polymer window.
Be Kα ×3.3
N Kα ×2.8
O Kα ×2.1
Si Kα ×1.5
Mn Kα ×1.4
www.microscopy-today.com • 2017 March
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76