7 # P ^2 O \ f \ f \f \f \f \t t \ \ \ \ \ ] ] ( ]6 x \ ] ] ] * ^) \f ] ] ] ] ^) ] ] ] ] ] ] LDEF IDE - Database: SIMS Analysis Procedure
SIMS Analysis Procedure
The procedure used for the SIMS analyses was continuously updated with more analyses being done. This description's intension is to give an overview of the common procedures. Part of it is based on the latest version of the procedure document at Analytical Instrumentation Facility, NC State University, which was authored, by Dieter Griffis and Charles Simon.
General
As the underlying priciple of SIMS is based on the analysis of surface material by sputtering this surface, and since all LDEF surfaces were covered with a surface contamination layer of generally unknown and varying thickness, this contanimation layer had to be removed first.
For each sensor, later for each feature except laboratory produced impacts, the contamination layer was analyzed by running a depth profile on a "clean" or "control" surface. SIMS datafiles in the archive having a filename and/or description that contains any of these two terms are no impact features.
Note that most of the difficulties encountered during the analyses were based on the
The goal of this experiment is to find out whether these craters originate from natural (micrometeorite impacts) sources or man-made (e.g. paint chips,etc.)
1D. SIMS analysis of IDE Sensors and other LDEF surfaces
The SIMS analysis may be broken down into four main portions as follows:
1. Preparation and loading of the sample and qualitative standard.
2. Depth profiling of contaminant layer.
3. Mass spectral analysis of remaining residue.
4. Secondary ion image acquisition
Sample and Standard Preparation
A qualitative standard has been prepared consisting of the following materials pressed into a ~5mm square piece of In foil: C, Al, TiO2, forsterite olivine (~20% MgO, ~75% SiO2, ~5% FeOx plus <1% Na. K, and Ca) stainless steel (~2% Mn, ~8% Ni, ~20% Cr, ~70% Fe), brass (Cu, Zn), Ag, and Au. This standard should be attached to the sample holder, or to the sample itself, using a piece of carbon tape. Try to keep the surface flat and as close as possible to the same level as the sample surface. You may use a piece of clean metal or glass to flatten the foil after it has been mounted. Check electrical continuity between the sample surface and the sample holder. Use the standard to identify element lines where indicated in the instructions below. Note the Hall probe setting when viewing the standard and again when viewing the sample. The setting has been observed to drift within the period of one analysis (<1 hour), therefore it is desirable to identify the element lines on the standard just prior to identifying the lines in the sample.
In some cases, noted below, unresolvable interferences will require the use of several isotopes and/or comparing 2-dimensional distributions (which must be non-coincident) for positive element identification. There will be some situations where positive identification is not possible, and these must be noted. each entry in the log book should have the image number, the mass, and detailed notes about the methods used to identify this mass as a particular species.
Before loading samples in the SIMS sample chamber, blow them off thoroughly with nitrogen or argon using the filter gun and look at them carefully in the optical microscope. Locate all of the features to be analyzed and then blow off the sample again just prior to loading it in the chamber. For IDE sensors, after the sample is loaded, check the electrical resistance between the sample surface and the Cameca sample holder to make sure it is <100 W. Reconnect the sample bias line (steel spring) from the sample stage to the high voltage feedthrough. After closing the sample chamber bulkhead, and before pumping down, make sure that the resistance between the center conductor of the sample bias feedthrough and the instrument housing is infinite. Locate the impact features to be analyzed through the Cameca optical microscope and record their micrometer values before evacuating the instrument . Also locate the In foil standard and assure that it is in the instrument's field of view. This is required in order to assure that a reasonable number of impact features (at least 2, preferably more) are within the field of view of the system. Fiducial marks will be made on the IDE sensors by C. Simon to assist in feature location.
Data Acquisition
Prior to any data acquisition, the conditions required for the analysis must be chosen based on crater size and the need for 3000 m/Dm mass resolution. Once these conditions have been chosen, they must not be altered during the analysis, i.e. the contrast diaphragm (CD) , the field aperture (FA), imaged field and raster size must remain constant throughout the analysis. CD 3 will always be used. Settings may not be changed during analysis and other conditions may NOT be used except after consultation with C. Simon or D. Griffis.
1. For this analysis, It is required that the instrument be operated at a mass resolution of 3000 m/Dm (except for very large craters, discussed below). This resolution requires the use of field aperture FA 2 for both the 150m and 400m image fields giving a 60m and 170m diameter imaged areas. For either of these imaging conditions, a 250m raster will be used. If FA 1, 400m imaged field is required based on crater size, a resolution of 3000 m/Dm is not practically acheivable. In this case, the entrance slit setting providing 3000 m/Dm resolution determined using FA2, 400m imaged field will be used. When using FA 1, resolution is dramatically affected by the raster size requiring the use of as small a raster as possible. Raster size to be used must be determined using the formula:
Raster sizeFA1/400Field = crater size + 50m + 1 beam diameter.
As discussed above, three possible spectrometer settings are chosen based on impact feature size:
Feature size 35 m
FA2, image field=150m dia., 250m raster, imaged area=60 m dia.
35 m 140 m (rare in our study)
FA1, image field=400m dia., choose a raster size based on the formula given above. Use of this formula will result in the smallest useful raster size (in 50m increments starting at 300m raster) and the maximum mass resolution.
2. Before beginning any data acquisition, use the MAIN program to locally calibrate to 27Al on IDE samples or a low Z, high intensity matrix element for other samples. Note the mass used for local calibration in the log. For local calibration to remain valid, the ESA voltages must be adjusted to their correct values (550.8, 509.1). Adjust these values if necessary before proceeding with the local calibration.
The CD, FA, image field and raster size determined based on the criteria above must be used throughout the analysis.
Depth Profiling
There is a ubiquitous layer of silicaceous/carbonaceous contamination on all LDEF surfaces. The thickness of this layer varies dramatically, even on a mm scale in specific instances, such as proximity to the dark rings around the electrodes on many IDE sensor surfaces. Auger analysis has shown that the thickness of the dark layer is about twice that of the lighter area just outside the ring. Frequently the differences in contamination layer thicknesses are visible, due to their C content. However, on surfaces that have been exposed to high levels of atomic oxygen, the C has been removed from the contamination layer and the differences in thicknesses are not visible.
The purpose of the depth profile (DP) is to determine the maximum thickness of this layer over an impact feature, and then to sputter through this layer before proceeding with the analysis. We are assuming that the contamination layer deposited on top of an impact site cannot be thicker than the contamination layer immediately adjacent (within ~1-2 mm) to the site. Operator discretion must be used in selecting the mass signals that will be used to determine when the contamination layer has been sputtered through. Note that the irregular surface topography will not permit complete removal of the contamination layer using these techniques. Our objective is to remove as little impactor residue as possible while removing as much of the contamination layer as possible. Note that DP's are no longer run on laboratory produced "blank" and "standard" impacts since we have shown that the contamination layer is blown away by the event.
1. Depth profiles are to be done first on an area adjacent to but not encroaching on the crater of interest. Elements listed below are to be profiled until C, Mg and Na reach a constant level. After the time necessary to reach these constant levels under the profile conditions described below has been determined, a DP run under identical conditions including identical time must be acquired on the crater site. Information specific to sample types is provided below.
>>>>>> Q.: Can we assume that contaminant layer uniformly covers sample surface? If yes -- we can use a single out-of-crater DP for a bunch of craters. If no -- we must run an out-of-crater DP for each and every crater.
IDE sensors: (Surface is composed of 0.1 m of PVD Al on top of 1 m of SiO2 on a Si wafer) Depth profile for 12C(O), 16O(-60), 23Na(-20), 24Mg(-60), 30Si(?), and 40Ca(O) looking for positive secondary ions: first run depth profile out of crater (but not very far away), then run DP on crater itself without Si. Note that Ca is a contaminant throughout the Al layer. We assume that the bulk of the contamination layer is removed as soon as the signals, especially C and Si away from the crater, level off. The DP over the impact site should be run under the same conditions for no longer than this amount of sputter time, or until the selected mass signals level off.
Other LDEF surfaces: Use the same procedure, but include Si in the DP over the crater on surfaces that are not composed of Si (Ge, Al and Au are the surfaces we have looked at to date in addition to the IDE sensors). Be aware that most micrometeorite residues will contain Si, so this signal cannot be used to determine when the contamination layer has been sputtered through over an impact site, but we do want to have a record of the Si signal during the DP.
2. DP's must be acquired using the FA, CD, image field and raster conditions selected for this analysis. Use a 500nA primary ion current, a fully open energy slit translated 2.5 turns, a -60 volt sample offset, and entrance and exit slits fully open.
3. Continue profile until C, Mg and Na intensities level off. This assures that the contaminant layer present on all LDEF surfaces is removed so that all data is taken on a surface with as little remaining contamination as possible.
4. Save both the file for the "off crater" and the "on crater" DP's and record the file names in the notebook along with relevant conditions (i.e. beam current and times required for contamination masses to become constant).
Mass Spectral Analysis (Bar Graph)
After completing the DP's described above, acquire bar graph spectra using the following instructions. Before acquisition of bargraphs, ESA and sample bias voltages must be checked and adjusted if necessary.
REM> Practice shows that 5 (five) nA is enough to get a good signal.
Choose Ip which gives 40Ca or 27Al reading of ~ 1..2e5 cts/s.
1. Conditions to be used for the bargraphs are the same CD, FA, image field and raster size determined above, a -60V offset and the energy slit fully open and translated 2.5 turns from the zero position.
2. Set the beam current to 5 na.
3. Check local calibration with MAIN program before proceeding. On IDE samples, using the MAIN program locally calibrate to 27Al. On other substrates, use Si or another low mass ion for local calibration. Then, if necessary reduce the beam current to adjust the intensity of 27Al on IDE samples or the brightest matrix ion on other samples (determined using MAIN) to <100,000 cts/sec for FA 2, 150m field, 250m raster or to <600,00 cts/sec for all 400m image field/FA combinations.
4. Acquire bargraph from 1-200 AMU. Print out the bargraph along with conditions, save file and record name of bargraph file in notebook along with relevant conditions.
5. Translate the energy slit so that maximum matrix intensity is obtained and set the sample energy offset to zero. Use the instructions in 3 above to set matrix ion intensity. Record a second bargraph and then repeat step 4 above.
Mass Resolution Adjustment
Mass resolution of 3000 m/Dm must be verified at mass 56 using the olivine standard and the raster size selected for the analysis (250m2 for FA 2, raster size determined by impact crater size for FA 1 based on formula given above) to measure the 56Si2/56Fe resolution ( m/Dm=2956). The slit settings required must always be noted in the log. Instrument must be warmed up for at least one hour prior to beginning high resolution tuning. Before determination of slit settings, ESA and sample bias voltages must be checked and adjusted if necessary.
1. Locate the olivine standard on the standard sample. Adjust the beam intensity to obtain an ion intensity of greater than 50,000 cts/sec at mass 56.
2. Fully close the energy slit and then open to 0.5 on the micrometer. Translate the energy slit to obtain the maximum counts per second.
3. Determine the entrance and exit slit settings required for a mass resolution (10% valley definition) of 3000 m/Dm.
4. Print out a copy of the mass scan obtained to verify resolution. On both the mass scan and in the log, record the entrance and exit slit settings required to obtain 3000 m/Dm.
5. Before proceeding and prior to exiting QB, all relevant instrument parameters must be checked, especially ESA and sample bias voltages. If ESA voltages require change, local calibration must be checked. Refocus image (check sample bias to make sure no residual energy aberrations are present by making sure image is uniform on channel plate).
Positive Secondary Ion Images
All images must be acquired using magnification setting 6 in order to provide maximum utilization of channel plate space. Make sure that the video camera is properly positioned and focused before starting analysis. Before image acquisition, ESA and sample bias voltages must be checked and adjusted if necessary.
1. Exit QUICKBASIC
2. Copy C:\CAMECA\COMMON.DAT to C:\RTI\MYDIR. COMMON.DAT: This file transfers local calibration data, its transfer to this directory is always required. Local calibration must be repeated if the polarity is changed or if ESA voltages require adjustment.
3. Increase primary beam current to 500 nA. Check the vacuum as indicated by the detector pump. At this point, check the vacuum as indicated by the detector and spectrometer pumps. If these vacuums are <1e-7 then turn off detector pump to reduce noise present on the channel plate. The detector ion pump must be turned on after each crater to insure that the channel plate is protected. If the vacuum does not quickly return to <1e-7, then the analysis must be terminated.
4. Check to see that the proper field aperature and image field are selected.
5. Run program SPACEACQ.EXE input filename ?????.img (need to add image extension).
6. For each mass, adjust the secondary magnet to the chosen mass. Enter the number of frames to be acquired. Adjust the mass before you give number of frames since adjustment is no longer possible once the frames have been entered.
Acquire images as follows: A comparison between the energy filtered and unfiltered bar graphs should be used to provide qualitative information on the degree to which molecular mass interferences are a problem. Be sure to record all image numbers next to the mass number and notes on species identification in the notebook. This is the only record that correlates the stored data with the analyses. Use the qualitative standard to identify element lines. Interferences are listed below along with comments on requirements for species identification. In general, a mass resolution (m/_m) requirement <2000 results in easily separable species. Required mass resolutions between 2000 and 3000 are achievable. Above 3000, special care must be used to resolve species and image quality is compromised. Species that require a mass resolution >3500 are generally not resolvable with our SIMS system and methodology. Many of the interference listed below are included for completeness even though they are very easily resolved. Some species have been left off this list, such as fractional ion species, B and Li compounds, since they are highly unlikely to be of concern in the LDEF samples. The energy-filtered bargraph mass spectra should be inspected carefully for indications of the expected (approximate) isotopic distributions for:
Mg 79%[24], 10%[25], 11%[26]
Si 92%[28], 5%[29}, 3%[30]
K 93%[39], 7%[41]
Ca 97%[40], 0.6%[42], 0.1%[43], 2%[44], 0.2%[48]
Ti 8%[46], 7%[47], 74%[48], 5%[49], 5%[50]
Cr 4%[50], 84%[52], 10%[53], 2%[54]
Fe 6%[54], 92%[56}, 2%[57], 0.3%[58]
Ni 68%[58], 26%[60], 1%[61], 4%[62], 1%[64]
Cu 69%[63], 31%[65]
Zn 49%[64], 28%[66], 4%[67], 19%[68], 0.6%[70]
Ag 52%[107],48%[109]
The presence or absence of the (more or less) appropriate isotope ratios for the elements above should be used whenever possible to determine the presence or absence of an element. While molecular interferences may skew the abundance ratios, the presence of all isotopes in the bargraph will indicate the possibility that the element species are present. The absence of all isotopes in the bargraph cannot be taken as conclusive evidence that the elemental species are not present deeper in the sample. In some cases it will be useful to record the images of interfering species. Operator discretion must be used here. The IDE samples, and other impactor residues, have a limited lifetime under the sputter beam.
In each case, an element line must be positively identified using the qualitative standard, non-coincident distributions with interfering species, isotopic abundances, or combinations of these techniques as required. The criteria utilized to make the identifications discussed above must be clearly described in the log book next to each image number n sufficient detail to allow the experiment to be reproduced. These details are critical to after-acquisition analysis.
Note that many interfering species, while resolvable from the element mass of interest, are not resolvable from each other. Acquire the masses in the order given.
Depending on the thickness of the Al layer you can acquire up to 1000 frames maximum for low intensity images. The image will go away due to charging if you have sputtered through the Al layer, if this occurs then note where this occurred and take no further images. Make sure you note in the log how many slit images you see for each mass and which slit image you recorded. ALWAYS use the standard to identify element lines and make note of this. This is the only way we will know if the images are the mass as noted above or if they are actually a molecular interference.
Consult appropriate Hall probe readings in the Log Book. Use 19-bit readings provided by the SPACEACQ program
On bright peaks reduce beam intensity by 1 order of magnitude. Keep the Channel Plate gain the same for all images in a set.
(We need to set a criteria as to how many frames to take based on intensity and need to preserve sample)
(The following sequence was suggested by yourself during the talk between three of us (you, me and Dieter)
Here's the sequence of acquiring the images:
1. 24Mg
2. 28Si2
3. 56Fe. If there is 56Fe -- proceed to 52Cr, 58Ni etc. (Cr and Ni are important if there is Fe.)
If NO 56Fe is present -- proceed to 47Ti.
4. All the rest.
5. 27Al
6. 28Si.
(New sequence -- 11/01/93):
12C
23Na
24Mg
27Al
30 Si for IDE
28Si- other sensors (non-Si based)
39K
40Ca
48Ti
52Cr
56Fe - both peaks
58Ni - both peaks - not resolved, cut with slit
61Ni - maybe: need to decide
63Zn
64Cu
107Ag
197Au
24Mg -- again
56Fe/Si2 -- both peaks
u <M< ΗOuc2NF, #$'02Yt×V_KAғJO ]PJEucc?%8#0K\9udoJ5ta {0hp& , , - - . . . ." .$ .( .* ., / / 0Q 0S 2 2 3b 3c 3 3 3 3 3 3 5 5 6
6, 8 8 : : H# I
@ V I I M N+ NE NG N N N N N N N N O O O O O O" O% O) O+ Oc Oi Ok Ox O O O O O O O O O O O O O O O O O O O O P P P P P P P P" P% P' P7 P9 Pg Pi P P P P P P P P P P P P P P P P
@ @ @ L < = D E f g 6 7 1 2 l m & X } ~
v w r s . / Y A B ] ^ n o " " $ $ % % , % , , ,
, L , , , , , , ,
, , , , E % (( () ) ) * * + + , , , , - - . .H .I / / /9 /: 1$ 1% 1 1 2 2 2 2 5 5 5 5 6+ 6, 6 6 7W 7X 8 8 8 8 : : :' :( ;* ;+ = = =Z =[ = = > > C C C D D' DX D D D D E E> ES ET , , , ,
, , Y , \ , ,
, , , , , u G ET H" H# I I J L MZ M M NE NF NG N N N N O^ O O O O O O O O O O O O P P P P P% P7 Pg P P P P P P P P P P P¹ª , , ! !$ ! !$ ! !$ ! !$ ! !$ , , , , , ,
, /
O P
P " . 8 C MG O n # U i I P ) * % ET P + , - 'T H H (FG( H H ( d ' @ =/ d R F H
-:LaserWriter
New York Courier ( (