TEM Sample Prep and Optical Microscopy
Refer to the original PDF for tables, figures and page layout.
Page 1\n\n\nPage 2\nIntroduction\n•\nMaterials Engineer\n•\nAllied High Tech Products, Inc.\n•\n4 years of laboratory experience\n•\nPrimary responsibilities:\n•\nProcedure development for \nvarious materials\n•\nImage Analysis \n•\nCustomer support/training\n•\nAllied High Tech Products, Inc.\n•\nFounded in 1983\n•\nAn American Manufacturer and \ndistributor of high quality \nequipment and consumables for \nmetallographic sample \npreparation and analysis.\n•\nZeiss Optical Microscope National \nDealer \n•\nMitutoyo Rockwell and MHT \nDealer\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 3\nTopics\n• Basics\n• What is a TEM sample?\n• Requirements?\n• Methods of preparing TEM \nsamples?\n• Mechanical preparation of a TEM \nsample\n• Preparation process\n• The role of optical microscopy in \nthe sample preparation process\n• Stopping point: Pre-FIB thinning \nor TEM Wedge?\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 4\nTEM – Transmission electron \nmicroscopy is a microscope technique in \nwhich a beam of electrons is \ntransmitted through a specimen to form \nan image.\nTEM Sample – Specimen that is usually \nultra thin (<100 nm) to allow electrons \nto transmit through.\nWhat is a TEM Sample?\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n[3] Zeiss HRTEM\n\n\nPage 5\nTEM Sample Requirements\nTEM samples must be electron transparent and generally the goal is a sample \nthickness of <100 nm\nAny deformation from thinning process or previous processing must be removed\nSmall area, < 3 mm or less, to fit various TEM grids and holders\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n[3] Specimen support grids of different \nmesh size and shape\n[3] Two-specimen holder with double-tilt\n\n\nPage 6\nCommon Methods of Preparation\n•\nIon Milling/FIB – Focused Ion Beam\n•\nProcess that involves bombarding a TEM \nsample with energetic ions or neutral atoms \nand sputtering material from the sample \nuntil it is thin enough to study in the TEM \n[3].\n•\nThe most versatile thinning process that can \nbe used for a wide variety of materials, \nhowever very expensive to initially invest in \nand to run.\n•\nDisadvantage is implantation of source \nmaterial and amorphous layer is created.\n•\nMost commonly used in semiconductors. \n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n[3] Schematic of a two-beam FIB and stages in \nmaking TEM samples using a FIB\n\n\nPage 7\nCommon Methods of Preparation\n•\nElectropolishing\n•\nSample is immersed in an electrolyte and \nthen subjected to a direct electrical current. \nThe sample is maintained anodic, with the \ncathodic connection being made to a nearby \nmetal conductor. [2]\n•\nAnodic dissolution of sample creates \npolished surface.\n•\nRelatively quick and can produce samples \nwith no mechanical damage, however can \nonly be used for electrically conducting \nmetals and alloys.\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n[3] The Window Method\n\n\nPage 8\nCommon Methods of Preparation\n•\nMechanical Preparation: Can be \nperformed by hand or with the aid of a \nmachine\n•\nManual: Utilizes hand tools (Tripods) that \nallow the user to make angle adjustments\n•\nSemi-Automatic: Precision polishers, such \nas the MultiPrep™ System, with digital \nindicators and dual micrometers allow: \n– Monitor amount of material removed in real \ntime in 1-micron increments\n– Precise sample tilt adjustments relative to \nthe abrasive plane. Alignment ensures \ncontrolled angles.\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 9\nMechanical Preparation\n• Advantages\n– TEM Wedge capabilities\n– All damage removed using the \n3x rule\n– No foreign material implanted\n– With experience it is repeatable \nand fast \n– Lower cost of required \nequipment\n– Decrease milling time and costs\n• Disadvantages\n– Requires careful handling \nthroughout process\n– Greater learning curve than others \ntechniques\n– Some samples may require one of \nthe other TEM preparation \nmethods such as metals due to \ninternal strain\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 10\nPreparation\n• Area of interest needs to be \nidentified; site-specific or bulk?\n• Once determined, sample needs \nto be prepared\n• Preparing TEM samples is \nessentially performing two \ncross-sections\nFirst side \ncross-section\nSecond side \ncross-section\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\nTarget\n\n\nPage 11\nThinning First Side\n• Sample placed onto a \ncross-sectioning paddle \nusing mounting wax\n• Paddle placed on hot plate to \nmelt wax\n• Light pressure can be applied \nusing a cotton-tipped applicator \nto assist in parallel registration \nof the sample to the fixture.\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 12\nThinning First Side\n• Stereomicroscope can greatly \nassist in aligning the sample to \nthe fixture\n• Tools such as the MultiPrep™ \nmay have methods of correcting \nfor the angle; however, using a \nstereomicroscope can minimize \nthe amount of corrections \nrequired\nTo paddle \nedge\nAlign sample \nfeatures\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 13\nThinning First Side\n• While the paddle is hot, it \ncan be taken to a \nstereomicroscope\n• This allows for sample \nadjustments before the \nwax cools and hardens\nInitial adjustments under microscope\nUse tweezers/tool to \nmake adjustments\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 14\nThinning First Side\n• Lapping films used as \nabrasive\n– Provide excellent edge \nretention and maintain \ncoplanarity\n– Typically used for \nunencapsulated cross-\nsectioning, TEM preparation, \nbackside polishing, etc.\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 15\nThinning First Side\n• Initial grinding with a coarse \nlapping film, such as 30 µm, will \ngive a ground edge\n• This ground edge can be used to \nobtain angle information using \noptical microscopy\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 16\nThinning First Side\n• Compound light microscopes \nwill give the resolution and \ndetails needed to perform \ncorrections\n• Imaging software can be used to \nobtain angle measurements\n88.87°\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 17\nThinning First Side\n• Angle corrections can be made \nusing the micrometers\n• Each tick mark is 0.02°\n• More material can be removed to \nobtain a new ground edge\n• With the new ground edge, the \nprocess can be repeated until an \nangle correction is no longer \nnecessary\n0.02°\nCorrected alignment\nPrevious alignment\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 18\nTarget: Site-Specific or Bulk?\nSite-Specific\n• Particular feature on \nsample\n• Failure Analysis\nMust stop at target, without \ndamage from previous steps\nBulk\n• Homogeneous material\nCan stop anywhere, however \nensure no damage from \nprevious steps\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\nTarget\n\n\nPage 19\n3X Rule\n•\nDamage is introduced into a sample \nthroughout the preparation process. \n•\nThis damage is not only represented by a \nscratch pattern on the surface, but also by \nsmaller micro-cracks that can propagate \nfurther into the sample that may not be \nseen by eye or optical microscopes \n(A 30 µm lapping film will not cause only\n30 µm deep scratches).\nSample\nDamage\n(micro-\ncracks)\n30 µm DLF\n15 µm DLF\nAOI\n9 µm DLF\n30 µm \nDLF\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 20\n3X Rule\n•\nIf a 30 µm lapping film is used and it is followed by a 15 µm lapping film step, the 15 µm DLF \nmust remove at least 90 µm of material to completely remove the damage from the 30 µm \nDLF. Similarly, if a 30 µm DLF is used and it is followed by a 6 µm DLF, then the 6 µm DLF \nmust remove at least 90 µm of material to completely remove the remaining damage.\n3X Rule\nCurrent Step\n(DLF)\nCurrent \nDistance from \nTarget\nPrevious Step\nRemove At \nLeast 3x Prev. \nStep Size\nDistance to \nTarget After \nCurrent Step\n30 µm\nVary\nN/A\nN/A\n192 µm\n15 µm\n192 µm\n30 µm\n90 µm\n102 µm\n9 µm\n102 µm\n15 µm\n45 µm\n57 µm\n6 µm\n57 µm\n9 µm\n27 µm\n30 µm\n3 µm\n30 µm\n6 µm\n18 µm\n12 µm\n1 µm\n12 µm\n3 µm\n9 µm\n3µm\n0.5 µm\n3 µm\n1 µm\n3 µm\nAt Target\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 21\n3X Rule\n•\nOther materials, such as ductile steels, \nmay only require a “2X Rule”, while \nfragile, brittle ceramics, where cracks \ncan propagate further into the sample, \nmay require a “4X Rule”.\n•\nThe main concept to understand is that \nno matter what steps follow, all damage \nintroduced by the previous abrasive \nmust be removed to get an accurate \nrepresentation on the microstructure.\nSample\nDamage\n(micro-\ncracks)\n30 µm DLF\n15 µm DLF\nAOI\n9 µm DLF\n30 µm \nDLF\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 22\nFlipping Sample\n• 2nd side needs to be prepared to \nthin sample\n• Previous method using cross-\nsectioning paddle can’t be used; \ncan’t overhang sample properly\n• TEM paddles with Pyrex® inserts \nare used to continue with the 2nd\nside\nPolished edge\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 23\nFlipping Sample\n• Surface of Pyrex® requires grinding\n– Grinding the surface of the Pyrex® will align it with that of the platen, \nensuring parallelism\n• Pyrex® can be ground using diamond lapping films\n– 6-3 µm finish is good enough\n– Do not want to polish the surface as having some scratches will help \nwith adhesion\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 24\nSite-Specific Orientation\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 25\nSite-Specific Orientation\n• With site-specific \norientation, the area of \ninterest is parallel to the \nPyrex® edge\nArea of Interest\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 26\nBulk Orientation\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 27\nBulk Orientation\n• With bulk orientation, the \narea of interest is \nperpendicular to the \nPyrex® edge\nArea of Interest\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 28\nFlipping Sample\n•\nSample waxed onto a glass slide\n•\n~1 mm thick piece sectioned from sample\n– This allows placement onto TEM paddle without risk of the sample breaking \noff due to being too tall\n•\nSample should be cleaned to remove abrasive particulates, excess wax, \netc.\n1 mm\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 29\nFlipping Sample\n• To ensure stability, a thin, \nuniform layer of wax needs to \nbe used.\n• Wax Bead Method\n– Place sample, polished side down, \ninto wax bead\n– Pick up sample and move over to \nempty part of glass slide\n– Allow wax to wick away\n– Repeat process 2-3 more times to \nend up with thin wax layer\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 30\nFlipping Sample\n• Polished side placed face down \nonto the Pyrex®\n• Positioned near the edge to \nallow easier thickness \nmeasurements later\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 31\nFlipping Sample\n• Stereomicroscope is great for \naligning to the edge of the \nPyrex®\nTweezers used to make \nslight adjustments\nSample edge flush with \nPyrex® edge\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 32\nFlipping Sample\n• Stereomicroscope is great for \naligning to the edge of the \nPyrex®\nTweezers used to make \nslight adjustments\nSample edge flush with \nPyrex® edge\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 33\nSecond Side\n•\nSecond side, whether site specific or not, now has a final target\n– Can lose sample if thin/polish for too long\n•\nChecking sample thickness becomes extremely important if you want \nto remove all the damage from previous steps\n•\nCertain machines, such as the MultiPrep™ System, may have \nindicators or other methods to track material removal\n– However, checking progress frequently with an optical microscope is a good way \nto verify material being removed: trust but verify\n1018 µm\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 34\nSecond Side\n• Mounting the sample close to \nthe edge allows for proper \nfocusing on the sample to \nmeasure side 1 edge\n• Compound scope can give \nrequired resolution to view \nsample thickness and make \nmeasurements\nPyrex®\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 35\nGoal: Pre-FIB or TEM Wedge?\nPre-FIB Thinning\n•\nSite-Specific Orientation\n•\nRemove the sample to prepare \nfor FIB thinning\n•\nSample can be removed by \nplacing paddle into container \nwith acetone (wax dissolves in \nacetone)\n•\nFilter paper placed around \npaddle so sample will eventually \npop-off and land on paper\nTEM Wedge\n•\nBulk Orientation\n•\nInduce an angle to create a wedge \nsample\n•\nAngle induced will vary depending \non the material being prepared\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\nThinned to electron \ntransparency\nThick enough for \nhandling\n\n\nPage 36\nInducing Wedge\n• Micrometers on MultiPrep™ used \nto induce an angle on the sample \nto create a wedge\n• Wedge allows for a sample thinned \non one edge, with a thick enough \nedge for handling\nParallel contact plane \n0.02°\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\nThinned to electron \ntransparency\nThick enough for \nhandling\n\n\nPage 37\nWedge Polishing\n•\nCertain TEM samples, along with \nthe use of a transmitted light \nmicroscope, can display a series of \ncolors in regions less than 10 µm \nthick\n•\nFringes can also occur in regions \nless than 2 µm thick\n•\nThe thickness that color \ncorresponds to will vary depending \non materials; however, some are \nwell documented, such as silicon\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 38\nWedge Polishing\nColor and fringe information can be \nused to assist in the preparation of \nTEM samples, as they are a guide to \noverall progress during a TEM wedge \nprocedure \n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n[4] Correspondence between Si wafer thickness and color of transmitted light\n\n\nPage 39\nWedge Polishing\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 40\nWedge Polishing\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 41\nWedge Polishing\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 42\nWedge Polishing\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 43\n•\n1. Allied High Tech Products, Inc., http://www.alliedhightech.com/. Accessed 23 May 2018.\n•\n2. Delstar Metal Finishing, Inc., Electropolishing A User’s Guide to Applications, Quality Standards \nand Specifications, https://www.delstar.com/assets/pdf/epusersguide.pdf. Accessed 23 May 2018.\n•\n3. Williams, David B, and C B. Carter. Transmission Electron Microscopy: A Textbook for Materials \nScience. New York: Plenum Press, 1996. Print.\n•\n4. Yougui Liao. Practical Electron Microscopy and Database. URL: \nhttp://www.globalsino.com/EM/page2805.html. GlobalSino 2007.\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n\n\nPage 44\nPablo Mendoza\nMaterials Engineer, Technical Services\nAllied High Tech Products, Inc.\npmendoza@alliedhightech.com\n© A L L I E D H I G H T E C H P R O D U C T S\nP A B L O M E N D O Z A\n