LabAdviser/314/Microscopy 314-307/Technique/X-ray spectroscopy/EDS-SEM: Difference between revisions
No edit summary |
|||
| Line 139: | Line 139: | ||
== 6) Select process time == | == 6) Select process time == | ||
The Process time (numbers from 1 – 6) is the length of time spent reducing noise from the X-ray signal coming from the ED detector during processing. By selecting different Process times, it is possible to reduce differing amounts of noise. The longer the Process time, the lower the noise. If noise is minimized, the resolution of the peak displayed in the spectrum is improved. In other words, the peak is narrower and it becomes easier to separate or resolve the peak from another peak that may be close by in energy. | |||
For the first look at a specimen, a long process time (5 or 6) should be used to start with in order to not missing any details in the spectrum. | |||
''Examples'' | |||
When identifying peaks closely spaced and overlapping, good peak separation is important. E.g. when looking at the overlap between Mo-L lines and S K-lines. | |||
Another example is the low energy K lines from light elements which is closely spaced with many peaks from L and M lines from elements with higher atomic number. When analyzing a sample at low HV to minimize the size of the interaction volume (looking at low energy lines), a good peak separation is necessary and can be achieved by using a high process time. | |||
[[File:Aztec-software_5.png|thumb|left|400px|Comparison of spectra Jadeite]] <br clear="all" /> | |||
When comparing spectrum 27 with spectrum 40 (especially for Al and Si), spectrum 27 has a better energy resolution than spectrum 40. However, spectrum 40 has almost ten times higher count rate. In spectrum 40, an Hg peak is observed, which is actually not present in the sample. This is due to a high count rate, which the system is not able to correct properly for pulse pile up, see below. | |||
When removing noise from the spectrum it will not only improve the peak separation but simultaneously limit the maximum acquisition rate, resulting in a “poor” counting statistic. For some measurements, analyzing for 5 or 30 seconds might not matter. However, when mapping, analyzing for 10 or 60 minutes matters. If mapping/analyzing an unknown sample, it is recommended to use a long process time in order to not miss any detail in the spectrum. However, if mapping certain elements whose main lines do not overlap, the process time can be decreased and the acquisition rate increased by increasing the beam current. | |||
The choice of process time depends on the sample. When lowering the process time, the dead time decreases, which means the beam current can be increased and thereby improving the counting rate. However, if the sample is charging, drifting or beam damage occur, a low beam current is necessary. When “forced” to use a low beam current, using a low process time as well does not make sense. It only makes sense when the beam current can be increased. | |||
[[File:Aztec-software_6.png|thumb|left|500px|Mapping of Jedeite]] <br clear="all" /> | |||
'''Pulse pile-up''' | |||
Pulse pile-up occurs when two X-rays pulses arrive almost simultaneously that the pulse processor cannot distinguish the signals as separate signals, which results in the two X-rays being measured as one. This is most commonly seen in the spectrum as a sum peak, which appears at the energy equal to the sum of two major peaks at lower energy. | |||
''Examples'' | |||
In the Figure ([[:File:Aztec-software_7.png|Pulse pile-up]]), a Hg peak is observed, which actually is a sum peak of O Kα and Si Kα. In the Aztec software, the spectrum can be viewed with and without pile-up correction. The green curve shows the spectrum before correction for pulse pile-up. The position of the Hg peak was corrected quit a lot, but the peak was not removed completely. | |||
Pulse pile-up can occur with any pair of pulses and continuum X-rays can pile-up with characteristic X-rays to produce a ‘pile-up continuum’ on the high energy side of major peaks. Pile-up of continuum X-rays with other continuum X-rays can cause the background to extend beyond the Duane Hunt limit or landing energy of incident electrons. The continuum artefacts can affect the accuracy of spectrum processing. Uncorrected pulse pile-up can cause a wide variety of analytical errors including element misidentification, inaccurate peak ratios and quantitative results which change with count rate. | |||
[[File:Aztec-software_7.png|thumb|left|400px|Example of pulse pile-up, where a Hg peak is observed as a sum peak of O K&alpha and Si K&beta. The green curve shows the spectrum before correction for pulse pile-up.]] <br clear=all> | |||
The next two figures show spectra of wollastonite (CaSiO<sub>3</sub>) both without and with pile-up correction, respectively. In the first figure ([[:file:File:Aztec-software_8.png|Spectrum without pile-up correction]]), the spectrum before pile-up correction shows a peak at ~5.5 keV (Cr Kα’s position), which is a sum peak of Si Kα and Ca Kα. In the second figure ([[:file:File:Aztec-software_9.png|Spectrum without pile-up correction]]), the spectrum after pile-up correction shows an over correction, where “holes” in the background is observed. | |||
<gallery mode="packed" widths="400px" perrow="1" halign="left"> image:Aztec-software_8.png|Spectrum without pile-up correction | |||
image:Aztec-software_9.png|Spectrum with pile-up correction </gallery> | |||
== 7) Bring the dead time to 30-50% == | |||
After choosing the process time, look at the dead time and adjust the beam current until it is 30–50%. | |||
'''Beam current''' <br/> | |||
You can increase the current by increasing the “spot” number, the current through the condenser lenses or by selecting a bigger aperture. | |||
'''Dead Time''' <br/> | |||
The percentage of time the pulse processor is unavailable for further counting. The ratio between input rate to acquisition rate. The input rate is the rate at which X-rays hits the detector. The acquisition rate is the rate at which the system is processing the counts and formulating the spectrum. Ideally the dead time should be somewhere between 30-70%. | |||
'''Too high current/dead time''': means probability of pile up is high, see above. | |||
'''Too low current/dead time''': means “bad” counting statistic. | |||
== 8) Aztec Software == | |||
Fill out the settings from left to right. Acquire image, spectrum. Process the data. | |||
[[File:Aztec-software.png|thumb|left|500px|Software flow]] <br clear=all> | |||
=== Point & ID === | |||
'''Qualitative, element identification''': <br/> | |||
In EDS, the qualitative analysis is the process of identifying elements present in a specimen. It involves acquiring a spectrum from the specimen and then identify the peaks in the spectrum. | |||
A step-by-step guide for qualitative analysis to get the most accurate results out of the system with minimal effort: | |||
Choose an appropriate high voltage, process time, dead time/beam current, see above. | |||
'''Describe specimen''': <br/> | |||
The specimen is described and if any knowledge of which elements are present, they can be pre-defined or auto ID is selected. | |||
'''Scan image''': <br/> | |||
Acquisition of an image, backscatter-, SE-image etc. Aztec uses the first quadrant of the microscope software. | |||
If the sample is drifting, use AutoLock. | |||
''AutoLock'' <br/> | |||
AutoLock is designed to increase the stability of data acquisition on SEMs and TEMs where the image may shift. This image shift can occur for a number of reasons, such as sample movement due to temperature changes at high magnifications or charging of the sample. | |||
It works by acquiring an image, comparing it with a reference image acquired at that Site, determining the image shift and adjusting the scan position to compensate for the shift. | |||
'''Acquire spectrum''' <br/> | |||
[[File:Aztec-software_10.png|thumb|left|300px|Acquisition parameters]] <br clear=all> | |||
Fill out the settings. For trace elements select a long acquisition time typically 10 to 30 seconds, for major elements a short acquisition time, typically 5 to 30 seconds, is selected. | |||
For process time, see above. | |||
A spectrum can be acquired from an area or point. | |||
'''Confirm elements''': <br/> | |||
Confirm which elements are present in the sample. A pre-knowledge about the sample is important if e.g. peak overlap or pile up peaks occur. | |||
To include or exclude an element, double click the element | |||
'''How to get results''': <br/> | |||
Select the Report Results and Report Template, where an appropriate template can be selected. It is possible to save the spectrum in different file formats by right clicking the spectrum. | |||
[[File:Aztec-software_11.png|thumb|left|300px|Results and report]] <br clear=all> | |||
'''Quantitative, standardless, normalized analysis''': <br/> | |||
=== Mapping === | |||
== 9) After use, retract the detector == | |||
Retract the EDS detector by using the control window found at the bottom right corner of the screen. | |||
[[File:Aztec-software_2.png|thumb|left|200px|Detector control]] | |||
:1) Select the button that looks like an EDS detector | |||
:2) Go to the “thermal” tab/window | |||
:3) Switch EDS from “operating” into “standby” mode | |||
:4) Go to the “Position” tab/window | |||
:5) Remove EDS detector from the chamber, by pressing the “Out” button | |||
The EDS detector retracts from the chamber | |||