Opt 307/407 Practical Scanning Electron Microscopy Considerations in any microscopy: Resolution Magnification Depth of field Secondary information.

Opt 307/407 Practical

Scanning Electron Microscopy

Considerations in any microscopy:Resolution

MagnificationDepth of field

Secondary information

Limits of Resolution (resolving power)

Unaided eye: 0.1mmLight microscope:0.2um

SEM: 1nmTEM: 0.2nm

Evolution of Resolution

Depth of Field

Light Microscope vs Electron Microscope

General Diagram of the SEM System

Light Microscopy vs Electron Microscopy

Advantages of EM:ResolutionMagnificationDepth of field

Disadvantages of EM:PriceyBetter if conductive (SEM)MaintenanceVacuum

Opt 307/407Vacuum Systems

Why do we need a vacuum anyway?

Electrons are scattered by gas (or any other) moleculesMFP at 1atm ~ 10cmMFP at 10-5T ~ 4m

Some samples react with gases (O2)

Helps keep things clean!


Terminology

PressureUnits: atm, bar, mbar

Torr (mm of Hg)Pa (N/m2)

1atm=1Bar=1000mBar=760Torr=105Pa

Pumping speedl/min, l/sec



Quality of Vacuum

Low: 760-10-2 Torr

Medium: 10-2-10-5 Torr

High: 10-5-10-8 Torr

Ultrahigh: <~10-8 Torr


Measuring Vacuum in EM Systems

Thermocouple GaugePirani Gauge

Cold cathode GaugePenning Gauge

Ion pump current

Very Broad Range of Vacuum to Measure

Grouped Ranges for Vacuum Gauges

Vacuum Gauge Choices and Working Ranges

Thermocouple/Pirani Gauges

Ionization Gauges

Ion Gauge Collection

Hot Cathode Ion Gauge

Penning gauge

Penning gauge



Types of Vacuum Pumps

1- Rotary (Fore, Rough, Aux, Mechanical)

2- Turbomolecular (Turbo)

3- Diffusion (Diff)

4- Ion (Sputter-ion)

Opt 307/407Vacuum SystemsRotary Pump Basics

Always in the Foreline of the system

Exhausts pumped gases to atmosphere

Pumping rate decreases as vacuum increases

Usually has a low VP oil as a sealant to facilitate pumping




Rotary Pump Problems

Cannot pump <10-2 TorrNoisy

BackstreamsVibration

Maintenance


Turbo Pump Basics

Direct drive electric motor-gas turbine

Rotor/stator assembly

Moves gas molecules through the assembly by sweeping them from one to another

High rotational speed (>10,000 RPM)

Very clean final vacuum

Opt 307/407Vacuum SystemsTurbo Pump Problems

Needs a Foreline pump

Costly

Can fail abruptly

Whine

Needs to be protected from solid material

Opt 307/407Vacuum SystemsDiffusion Pump Basics

No moving partsHeated oil bath and condensing chamberJet assembly to redirect condensing gas

Recycle of oil

Pressure gradient in condensing chamber/Foreline pump removes from high

pressure side



Diffusion pump problems

Heat up/cool down time

Needs foreline pump

Can make a mess in vacuum failures/overheating

Needs cooling water (usually)


Ion Pump Basics

High voltage creates electron fluxIonizes gas molecules

Ions swept to titanium pole by magnetic fieldTitanium erodes (sputters) as ions become

embedded

Getters collect Ti atoms and more gas ionsCurrent flow indicates gas pressure (vacuum)



Ion Pump Problems

Cannot work until pressure is <10-5 Torr

Low capacity storage-type pump

Needs periodic bake-out

Hard to startup (sometimes)


Summary

All electron microscopes require a vacuum system.Usually consists of rotary-(turbo, diff)-(ion) pumps.

System should provide clean oil-free vacuumat least 10-5 Torr or so.

Vacuum is usually measured with a combination of TC and ion gauges.

Vacuum problems are some of the most challenging to find and fix, and may even be caused by samples

outgassing



Typical TEMVacuum System

Opt307/407

Electron Sources and LensesElectron Sources and Lenses

Types of Electron SourcesTypes of Electron Sources

Thermionic SourcesTungsten filamentLanthanum Hexaboride (LaB

6) filament

CeB6

Field Emission sourcesColdSchottky

Ideal Electron Source CharacteristicsIdeal Electron Source Characteristics

Low “work function” material so that it is easy toremove electrons from the material

High melting point

Chemically and physically stable at high temps

Low vapor pressure

Rugged

Cheap

Thermionic Emission of ElectronsThermionic Emission of Electrons

Filament material is heated with an electrical currentso that the “work function” of the material is exceededand the electrons are allowed to leave the outermost orbital.

Generates a fairly broad source of electrons (cloud)

Tungsten Hairpin FilamentsTungsten Hairpin Filaments

Most common of all filaments in electron guns

Low cost (~$20)

Lots of beam current

Not very intense illumination

Emission temperature ~2700K

Work function= 4.5ev

Can last about 100 hours

Tungsten Hairpin Filament SaturationTungsten Hairpin Filament Saturation

Tungsten Hairpin FilamentTungsten Hairpin Filament

LaBLaB66 (and CeB (and CeB66) Filaments) Filaments

Lower work function thermionic source (2.4ev)

Lots brighter (~50x) than W-hairpin

Relatively costly (~$700)

Can be direct replacement for W-hairpin

Heated to about 1700K

Can last hundreds of hours

LaBLaB66 Emitter Problems Emitter Problems

Need higher vacuum to reduce reactivity

More difficult to make

Heating/cooling must be slow (brittle material)

Heating is indirect through a graphite well

Thermionic Gun LayoutThermionic Gun Layout

Optimization of Thermionic Emitter LifetimeOptimization of Thermionic Emitter Lifetime

Keep vacuum system in good working order

Clean gun area

Do not oversaturate the filament

Minimize the number of heating/cooling cycles

Field Emission Electron SourcesField Emission Electron Sources

Process proposed in 1954/Demonstrated in 1966

Usually a single crystal W-wire sharpened and shaped

Tip radius <1.0um

Usually includes a ZrO2 component to assist emission (if heated)

About 10,000 times brighter than W-hairpin

Small apparent source which helps obtain small probes with high temporal coherence

Decreased energy spread in the beam

Can last many thousands of hours

Cold Field EmittersCold Field Emitters

Most intense (brightest) electron source

Tip radius very small (~0.1um)

Needs very high electric field intensity

Tips contaminate and need “flashing” to clean and/or anneal

Expensive (~$4000)

Requires ultrahigh vacuum in gun

Schottky Field EmittersSchottky Field Emitters

More stable than cold field emitters

Self annealling as ions impact tip

Lower work function than cold field emitters

Extraction field intensity can be lower

Vacuum requirements lower

Still expensive (~$4000)

Typical Schottky Field Emission SourcesTypical Schottky Field Emission Sources

Schottky Field Emitter DiagramSchottky Field Emitter Diagram

Suppressor Cap:limits the electron emission to the desired area of the tipactually blocks electrons from the heater and shaft

Heating Filament-tungsten hairpin:heats the tungsten tip to enhance emission (1800K)

Emitter:Single crystal W-needle w/ ZrO2 coating

Schottky Field Emitter PartsSchottky Field Emitter Parts

Extractor Anode:applies voltage to the filament to extract electrons from the tip (1.8 - 7 keV)

Gun Lens:Electrostatic lens which forms a crossover of the electron source (acts similar to the C1 lens)

Schottky Field Emitter PartsSchottky Field Emitter Parts

Optimizing Field Emission Emitter LifetimeOptimizing Field Emission Emitter Lifetime

Keep vacuum system in good working order

Leave the emitter heated

Don’t over-extract

Don’t overheat

Electron LensesElectron Lenses

ElectrostaticGun cap (Wehnelt cylinder)Totally inside vacuum

ElectromagneticAll other lenses and stigmatorsPartially outside of vacuum

Transmission Electron MicroscopeTransmission Electron MicroscopeOptical instrument in that it uses a lens to

form an image

Scanning Electron MicroscopeScanning Electron MicroscopeNot an optical instrument (no image forming

lens) but uses electron optics. Probe forming-Signal detecting device.

Electron OpticsElectron Optics

Refraction, or bending, of a beam of illumination is caused when the ray enters a medium of a different optical density.

Electron Optics

In light optics this is accomplished when awavelength of light moves from air into glassIn EM there is only a vacuum with an optical density of 1.0 whereas glass is much higher

Electron Optics

In electron optics the beam cannot enter a conventional lens of a different refractive index. Instead a “force” must be applied that has the same effect of causing the beam of illumination to bend.

Classical optics: The refractive index changes abruptly at a surface and is constant between the surfaces. The refraction of light at surfaces separating media of different refractive indices makes it possible to construct imaging lenses. Glass surfaces can be shaped.

Electron optics: Here, changes in the “refractive index” are gradual so rays are continuous curves rather than broken straight lines. Refraction of electrons must be accomplished by fields in space around charged electrodes or solenoids, and these fields can assume only certain distributions consistent with field theory.

Converging (positive) lens: bends rays toward the axis. It has a positive focal length. Forms a real inverted image of an object placed to the left of the first focal point and an erect virtual image of an object placed between the first focal point and the lens.

Diverging (negative) lens: bends the light rays away from the axis. It has a negative focal length. An object placed anywhere to the left of a diverging lens results in an erect virtual image. It is not possible to construct a negative magnetic lens although negative electrostatic lenses can be made

Electron OpticsElectrostatic lens

Must have very clean and high vacuum environment to avoid arcing across plates

Electromagnetic Lens

Passing a current through a single coil of wire will produce a strong magnetic field in the center of the coil

Three Electromagnetic LensesThree Electromagnetic Lenses


Pole Pieces of ironconcentrate lines ofmagnetic force



Forces Acting on an Electron Beam as Forces Acting on an Electron Beam as it goes through an Electromagnetic Lensit goes through an Electromagnetic Lens

...and the Result...and the Result

The two force vectors, one in the direction of the electron trajectory and the other perpendicular to it, causes the electrons to move through the magnetic field in a helical manner.

The strength of the magnetic field is determined by the number of wraps of the wire and the amount of current passing through the wire. A value of zero current (weak lens) would have an infinitely long focal length while a large amount of current (strong lens) would have a short focal length.

Condenser Lens: Weak and Strong ConditionsCondenser Lens: Weak and Strong Conditions

Lens DefectsSince the focal length f of a lens is dependent on the strength of the lens, if follows that different wavelengths will be focused to different positions. ChromaticChromatic aberration of a lens is seen as fringes around the image due to a “zone” of focus.

Lens DefectsIn light optics wavelengths of higher energy (blue) are bent more strongly and have a shorter focal length

In the electron microscope the exact opposite is true in that higher energy wavelengths are less effected and have a longer focal length

Lens Defects

In light optics chromatic aberration can be corrected by combining a converging lens with a diverging lens. This is known as a “doublet” lens

The simplest way to correct for chromatic aberration is to use illumination of a single wavelength! This is accomplished in an EM by having a very stable acceleration voltage. If the e velocity is stable the illumination source is monochromatic. monochromatic.

Lens Defects

A few manufacturers have combined an electromagnetic (converging) lens with an electrostatic (diverging) lens to create an achromaticachromatic lens

LEO Gemini Lens

The effects of chromatic aberration are most profound at the edges of the lens, so by placing an aperture immediately after the specimen chromatic aberration is reduced along with increasing contrast

Lens Defects

The fact that rays enter and leave the lens field at different angles results in a defect known as sphericalspherical aberration. The result is similar to that of chromatic aberration in that rays are brought to different focal points

Spherical aberrations are worst at the periphery of a lens, so again a small opening aperture that cuts off the most offensive part of the lens is the best way to reduce the effect.

Diffraction

Diffraction occurs when a wavefront encounters an edge of an object. This results in the establishment of new wavefronts

Diffraction

When this occurs at the edges of an aperture the diffracted waves tend to spread out the focus rather

than concentrate them. This results in a decrease in resolution, the effect becoming more pronounced with ever smaller apertures.

AperturesApertures

AdvantagesAdvantages

Increase contrast by blocking scattered electrons

Decrease effects of chromatic and spherical aberration by cutting off edges of a lens

DisadvantagesDisadvantages

Decrease resolution due to effects of diffraction

Decrease resolution by reducing half angle of illumination

Decrease illumination by blocking scattered electrons

If a lens is not completely symmetrical objects will be focussed to different focal planes resulting in an astigmaticastigmatic image

Astigmatism

The result is a distorted image. This can best be prevented by having a near perfect lens, but other defects such as dirt on an aperture etc. can cause astigmatism

Astigmatism in light optics is corrected by making a lens with a offsetting defect to correct for the defect in another lens.

In EM it is corrected using a stigmator which is a ring of electromagnets positioned around the beam to “push” and “pull” the beam to make it more circular in cross-section

Opt 307/407

The SEM Systemand

Electron Beam-Sample Interactions

The TEM system and components:

Vacuum Subsystem

Electron Gun Subsystem

Electron Lens Subsystem

Sample Stage

More Electron Lenses

Viewing Screen w/scintillator

Camera Chamber

The SEM System and Components:

Vacuum Subsystem

Electron Gun Subsystem

Electron Lens Subsystem

Scan Generator Subsystem

Scattered Signal Detectors

Observation CRT Display

Camera CRT/Digital Image Store

SEM Scan Generation System

Sets up beam sweep voltage ramp in both X and Y directions (tells beam how far to move and the number of increments)

Synchronized between beam on sample and beam on CRT display

Can be analog or digital in format

Includes interface to magnification module for changing the beam sweep on the sample

Scan Generator Interface

Magnification control in the SEM

Beam sweep on sample is synchronized with beam sweepon display CRT

CRT size never changes

Sweep distance on sample can vary (using magnificationmodule)

Small distance on sample--> large magnification to CRTLarge distance on sample--> small magnification to CRT

Mag=CRT size/Raster Size

Magnification Control in the SEM

Depth of Field in the SEM

The single most important thing in making SEM imagespleasing to look at and interpret

Range of distances above and below the optimal focusof the final lens that produces acceptably focussed imagefeatures

DOF in the SEM is a few hundred times that of the LMat similar magnifications

DOF is inversely proportional to the aperture angle

Depth of Field and Defocus

DOF in the SEM

DOF and Aperture Size

Table 1. Depth of Field at 10 mm working distance for SE images.

Magnification

30 1.9 mm

3000

30000

Table 2. Depth of Field at 25 mm working distance for SE images.

Magnification

30 4.9 mm 2.5 mm 1.6 mm

3000

100 m aperture

200 m aperture

300 m aperture

( = 0.005 rad) ( = 0.01 ( = 0.015

995 m 663 m

10 m 5 m 3 m

1 m 0.5 m 0.3 m

100 m aperture

200 m aperture

300 m aperture

( = 0.002 rad) ( = 0.004 rad)

( = 0.006 rad)

25 m 12.5 m 8.3 m

Note the large depth of field which is possible with small probe semi-angle ( .

DOF and Sample Tilt

DOF and Working Distance

Spot Size

Resolution is a direct function of (and limited by) the final spot size of the electron beam

This is a function of initial beam crossover size at the gunand the final spot formed by the beam shaping apertures and the condensing lenses

Shorter focal lengths produce smaller focussed spots

Short working distances have the smallest spots andthe best resolution

Smaller spots reduce the signals generated (S/N decreases)

Spot Size Control in the SEM

Signal Detectors for the SEM

Electron Beam-Specimen Interactions

First thing: electrons are scattered in a near-forward direction

Electron Beam-Sample Interaction

Electron Flight Simulator Demo

Smorgasbord of Electron Beam Sample Interactions

Elastic ScatteringBackscattered Electrons

Inelastic ScatteringPlasmon Excitation (coherent oscillations in free electron “plasma”)

Secondary Electrons from conduction band

Electron Shell Excitation (photons, characteristic x-rays and Auger electrons)

X-ray Continuum (braking radiation)

Phonon Excitation (thermal)

Electron Beam-Sample Interactions

Backscattered (Primary) Electrons

Backscatter Yield

n=-0.0254+0.016*A2-0.000186*A2*A2+0.00000083*A2*A2*A2

Backscatter Yield

0

0.1

0.2

0.3

0.4

0.5

0.6

0 20 40 60 80 100

Atomic #

Yie

ld

Backscattered Electron Detectors

Backscattered Electron Image

Backscattered Electron Detector Placement

For either solid-state Si detectors or Robinson type

Secondary Electrons and Detectors

Secondary Electrons

Inelastic collision and ejection of weakly held conductionband electrons (need only few eV to exceed work functionof the sample atoms)

Always low in energy (<50eV)

Can also be formed from backscattered electrons. Ratio is Zdependent (SE

BS/SE

B increases with Z)

Usually a large fraction is produced within a region definedby the primary beam

Some Secondary Electron Characteristics

Types of Secondary Electrons/Origins

Secondary Electrons: Edge Effects

Everhart-Thornley (ET) Secondary Electron Detector

Photomultiplier Tube Electronics

Whole E-T Detector w/PMT Amplification

Secondary Electron Images

Auger Electron Generation

Auger Analytical Volume

Auger Electron Spectroscopy

Yielded inverse to BSE: lighter elements emit more

Electrons are VERY specific in energy...can indicatetype of bonding involved and oxidation state

MFP for typical Auger energies is about 0.1-2nm

Analytical volume is very small---> resolution is high

Signal is pretty weak

X-ray Photon Production

Bremsstrahlung (Braking) radiation

Characteristic X-rays

Bremsstrahlung Continuum X-rays

Formed by the release of energy from the primary electronbeam as it decelerates in the presence of the Coulombicfield of target (sample) atoms

Large energy spread (0-E0)

Not very useful

Forms a large portion of the x-ray spectral background

Characteristic X-rays

Formed when inner shell electrons are ejected by the primary beam, followed by an outer shell electron

falling and filling the vacancy. Energy difference iscompensated by releasing a photon of “characteristic”energy, defined by the energy level differences of the orbitals, which is unique within a series of transitions.

Characteristic X-ray Production

Energy Dispersive X-ray Spectrometer

X-ray Spectrum from EDS Spectrometer

Wavelength Dispersive (crystal) Spectrometer

X-ray Spectra Comparison EDS vs WDS

Cathodoluminescence Signal Generation

Electron beam excitation of sample valence band electronsinto the conduction band (electron-hole pair production)

If allowed to recombine, the annihilation of the electron-hole paircreates a photon (sometimes in the visible range)

A high efficiency collector (usually a parabolic mirror) and aPMT are used to collect and amplify the signal

Absorbed Current or Specimen Current

Sample is detector

IB~= I

SC+ I

BS + (I

SE + I

ph +I

etc)

SC image looks like an inverted BSE image

Very useful and easy to obtain

Resolution not so good

Transmitted Electrons

In thin samples the beam may pass through the thickness

TED is located below the sample (like BSE detectors)

Sort of like TEM w/o the resolution

Relative Sizes of the Emission Zones (looking from above)

Image Collection, Recording and Presentation

Rule-of-thumb microscope conditions-best resolution-best depth of field-best sample preservation

Conventional Photographic Methods

Digital Methods

Presentation for:DisplayPublication

Image Collection

Proper subject identification

Proper subject orientation

Best selection of imaging conditions-HV-WD

-Spot size (aperture)-Scan rate

Subject Identification/Orientation

Representative of the whole

Image background

Not too busy

Important image information is centered and prominent

Many times a slight tilt conveys more information

“Best” Imaging Conditions

High resolution-short working distance

-small spot size-high accel. Voltage-high magnifications

Depth of field-long working distance

-low magnifications-larger spot size

Low magnification-large spot

Selection of Scan Rate for Imaging

Sensitive samples-may need to be fast

-low S/N-maybe TV integration mode

Insulating (charging) samples-decrease charging with small spot and

fast frame rate, maybe TV again-focus/stigmate in an area adjacent to the area recorded

-use image shift function to quickly move small amounts

Normally conductive samples-use slowest rate practical w/o degrading surface

Old Technology

Analog scan SEMs

2nd CRT for viewing the image as it scans

Film based camera focused on this CRT (low persistence)

Almost always a 4x5 inch Polaroid sheet film camera

Very slow scan for about a 2000 line image (~3 minutes)

P/N film or just an instant positive image

About $3/shot now

Generalized Photographic Processing

Needed for TEM image plates(Can be used for SEM film images too)

Exposure of silver halide grains (latent image)Development (reducing basic solution---> Ag0)Rinse (water) or Stop (acid)Fix (thiosulfates)Rinse (water)Dry

Scan or Print photographically

Good photographic processing results in the best imagesand are still the images that are used to compareother (newer) techniques

Newer Technology

Digital raster SEMs

Frame buffer storage of image info

Image processing

Digital image storage-usually TIFF files so that header can contain

image and microscope specific data

Fully transportable formats

Easy incorporation of images into documents

LEO 982 Specific Digital Imaging

Detectors-SEI (chamber)-SEI (column)-BSE

Signal mixer-brightness-ratio

Gamma correction-corrects for desired brightness and contrast I

out~=I

in

-power function deviation from 1:11.0 darkens and enhances lower greys1.0 lightens and enhances higher greys

<--- switch position 0


Gamma Corrections

<----- switch position 3




Gamma corrections


Slow scan rates 1-3 continuous scan

Slow scan rates 4-8 store one frame of data-dump to disk as image file (TIFF)

Choose image pixel matrix density from 512x512 to2048x2048 (lowest is usually OK)

Right mouse button will interrupt any scan and storeresults in the buffer (incl. TV)

TV rate integration of frames can reduce random noisein the final image at a fast scan rate

File path and naming convention


Variable small raster-used to increase scan rate for image adjustment

Can store multiple images in the same frame-variable frame

-split screen-kind of gimmicky.....don't use for important images

Stereo Pair Images (Anaglyphs)

By collecting two images offset by about 4-100 in tilt

Display them side by side and cross eyes to converge

Build a blue-red image composite and use stereo glasses-In Photoimpact program:

convert images to RGBadjust color balance (red-right, blue-left)

perform image calculation (difference operator and merge)

Special Scan Modes in the LEO 982

Line scan-disable Y-axis scan to see grey-level variations

on a line

Y-modulation-if very little Z-axis information this converts it

to Y-axis deflection (not very useful)

Spot scan-mostly for x-ray data acquisition

Additional Scanning Features of the LEO 982

Dual magnification-useful for “looking around”-don't use for important images

Scan rotation-electronically rotates the raster on the sample-very useful for getting a good “presentation”

Dynamic focus-use to compensate for the portions of the sample that

fall outside the depth of field distance. Sets up aramp on the focus current +- the center of the field

Tilt correction-compensates for trapezoidal scan on highly tilted samples

Image Processing

Generally use “kernels” which are arrays of arithmeticoperators on a pixel

Standard kernels are used to blur, average, and sharpenimages. 3X3, 5x5, array of operators.

Photoshop and PhotoImpact have custom and standardkernels

Kernel Operations for Sharpening an Image

Different Kernels

Effect of Kernel Size on Operations

Contrast Enhancement

Original kernel Average kernel

Sharpen kernel Blur kernel

Pitfalls of Image Processing

Images can be distorted and data lost

Pixelization of images

Ethical behavior dictates a minimum of processing

Always better off collecting the best image and eithernot processing or doing it only lightly

Image Manipulation

Erosion of edge pixels-kernel operator to find edges

-erode or erase edge pixels one layer at a time-break apart and separate touching features

Dilation of edge pixels-kernel operator to find edges

-dilate or add edge pixels one layer at a time-fuse separate features

Most useful in particle and other small repeating features

Presentation of Micrographs

Reports-probably least critical

-must convey information concisely

Journal-probably most critical

-size, grey-levels, resolution-must be specific and representative of the narrative

Posters-most variable in format-otherwise like journal

-conducive to point and discuss

Web-like journal

-can be interactive

Presentation Media

Photographic paper

Photo quality printer output-dye sublimation

-ink jet....getting there!-laser...maybe...

-consider viewing distance in choice

Include TIFF or JPEG files in reports using word processor

Powerpoint for talks

Micrographs as Art

Wonders of things small

Intricacies of natural samples

Subtle grey tones, like fine b/w photos

Can be psuedocolored to add interest

Comparisons to more familiar things

Explain phenomena in a “gee-whiz” way

Sample Preparation for Electron Microscopy

Electrically Conductive SamplesElectrically Insulating SamplesBiological Samples“Odd” Samples

Why do samples need to be prepared???

Vacuum environment

Charged particle environment

Too big

Components migrate in response to the beam

Two General Samples Types

Bulk SamplesSEM only

Thin SamplesSEM and TEM

Processes Common to Many Samples

Dehydration

Coating

Methods to reveal interior details

Stabilization of loose parts

Sample resizing

Methods to make similar measurements with other techniques

Special imaging circumstances

Dehydration

Why? Samples are incompatible with the vacuum

Surfaces will be disrupted while forced-drying

How?Air dry

Critical Point DryHMDS Dry

What sample types?Biologicals

Hydrated geologicalsSynthetics like polymers or solgels/aerogels

Air Drying

Can only be used on “rugged” samples

Biologicals like tough exoskeletons

Materials that won't change size/shape

Air Dried Sample

Critical Point Drying

Water is replaced with miscible 2nd fluid

Transitional fluid replaces 2nd fluid

Transitional fluid is driven past the “critical point”by increasing pressure and temperature

Pressure is relieved as gas escapes

Samples are left water, 2nd fluid, and transitional fluid dry

CPD Sample

Critical Point Dryer

More CPD Dried

HMDS drying

Water is replaced with a 2nd fluid

2nd fluid is replaced with HMDS

HMDS is allowed to dry leaving surfaces intact

HMDS Dried


Dehydration

Coating



Sample resizing



Sample Coating

Why coat samples?Electrical insulators need to be made conductive

Increase rigidityIncrease SE emission

Usual coatingsMetals like Au, Ag, Pt, Pd, Cr, Os or alloys

Carbon

Typical coating methodsSputtering

Evaporation

Sample Coating

Things to watch out for:

Decoration artifactsX-ray emission lines

Sample deformation during deposition

Sputter Coating Samples

Usually a simple DC sputtering system

Low vacuum

Argon backfillinert and ionizable

relatively high massgood pumping character

Relatively simple time vs current rate of deposition

Slower coating--->smaller islands--->smoother film

Usually +-5nm is sufficient for conductivity

Typical EM Lab Sputtering System

Cathode

Vacuum chamber

Samples

Vacuum gauge

HV control

Current monitor

Timer

Argon bleed

Sample Coating: Evaporation

Used when sputtering won't work wellCarbon

Making shadows

Line of sight deposition


Dehydration

Coating



Sample resizing



Revealing Interior Portions of Samples

Why?Outside may be “weathered”

Inside may have different chemistry or morphology

Inside may have smaller pieces or details

Inside may be immature or undifferentiated

Inside may be source of problems or defects


How?Smash it! (don't make it any harder than necessary)

Cut it

Saw it

Grind it

Fracture it

Polish it (mechanical, electrochemical)

Etch it


Toolsvarious types of knives and blades

Microtome

Polishing bench and wheels

Wet processing

Inside Structure

Microtomes and Microtomy

Tool with very sharp blade and a sample translation stage

Ultramicrotome for EM

Usually a glass or diamond knifestationary cutting edge

moving samplecut pieces float off on water surface held adjacent

to the blade edge

Can use thin sections in TEM or cleaned bulk surfacein the SEM


Dehydration

Coating



Sample resizing




Why?Loose stuff falls offLoose stuff changes other surface details

How?Use glues or tapesUse clipsMake sandwichesEmbed in other materialsSometimes a coating will do


Dehydration

Coating



Sample resizing



Sample Resizing

Why?Too darned big for the system

How?Similar to revealing interiors of samples

-smash, saw, cut, grind, polish, etc.

Concerns:Part left over is representative of the whole

You don't lose the interesting part


Dehydration

Coating



Sample resizing




Why?Complementary data

Comparisons

How?Use fiducial markings

Use sample holders with a grid of numbers/lettersFind a landmark

Use absolute or relative stage coordinatesCircle the area of interest


Dehydration

Coating



Sample resizing




Why?Want sample in particular positionNeed to see a certain area or side

Want proximity data to/from reference material

How?Be creative

Mount samples so they protrude from stageMake a multi-holder

Include a standard material on the stageSpring clips/tape/wire

Sample Preparation Flowchart

Individual Processing of Samples for EM Observation

Small ParticlesCross-sectionsInsulatorsConductorsBiologicals“Untouchables”For automated analysesSemiconductor devicesManipulated samplesHigh or Low temperature processingLow vacuum observationHazardous materials“Quick-and-dirty” analyses

How to Prepare Small Particles

Dispersion of single particles or groupings?

Mixture of sizes or monodisperse?

Potential to move around on stage?

Want compositional information? What about the substrate?

From a solid mass, dry powder, airborne, or liquidborne?

Reactive outside of their usual environment?

Small Particle Dispersion

Agglomeration is a problem-camphor/napthalene method-sticky dot method-dust and remove method-filter onto membranes (Nuclepore filters)

Drying ring dispersions

Mortar and pestle size modification

Small Particles

Most will stick electrostatically

Large ones may need some help to stay in place-carbon coating-metal coating

-sticky dots

Coatings often are not continuous-special stages for evaporators and sputter coaters



Cross Sections

Why?-to see interior or sub-surface details

How?-fracture-cleaving

-microtome-polishing



Electrically Insulating Materials

Four Choices:Try to view as-is w/low energy beam

-small aperture-vary accelerating voltage

Try a faster scan rate to limit electron dose

Make it conductive w/o destroying thesurface topography

Use a variable pressure instrument (we don't have one)

Insulators



Electrically Conductive Samples

The best sample

The most unusual sample

Simply attach to sample stub and “go”

Beware of contaminated surfaces



Biological Materials

Generally require extensive preparation

Most important to remove water w/odestroying the surfaces

May need to ruggedize (fix) tissues

May be possible to freeze and view directly

Given rise to “environmental” or “low vacuum”systems to obviate need to dry samples



Untouchable Samples

Historically significant samples

Forensic samples

Samples from litigations



Preparing Samples for Automated SEM Scans

Usually a size/shape/compositional analysis

Usually requires a grey-level segmentation of the image

Usually needs some parameters to keep or discard data-edges

-too small-too big

Samples must be flat and relatively featureless except for your target

Examples:gunshot residue analysis

asbestos analysisbone implant analysis

small particle analysis (IPA, SPOT sampler)

Gunshot Residue Analysis

When a gun is fired, small particles are generated during the explosion of the primer,

and leave the gun via the smoke.

The particles are deposited on parts of the body.

These small particles are called gunshot residue (GSR).

Particles are very characteristic, therefore presence of these particles forms evidence of firing a gun.

Particles normally consist of Pb (lead), Sb (antimony) and Ba (barium).

Gunshot Residue



Sample Preparation of Semiconductors

Usually Silicon

Increasingly III-V or II-VI compounds

Do not need conductive coatings unless a thick oxide,nitride or resist is present

p-type and n-type seem to image differently due tovariation in conductivity and dopant concentration

Some areas may be “floating” electrically and needseparate grounding



Manipulated Samples

Stressed in tension or compression

Samples irradiated to simulate high dose -exposure

Electron beam induced current (EBIC)

Voltage contrast



Temperature Controlled Viewing in the SEM

Some glasses have mobile components-Na+-Ag+

Cooling to <-140C seems to stabilize the electromigration

Some high VP or liquid samples can be frozen and viewedw/o a coating

Watch the crystallization of materials from solution



Low Vacuum SEM

ESEM (environmental SEM)

Differentially pumped gun/column and chamber

High vacuum in former; adjustable vacuum in latter

Many types of backfill gasses and vapors

Up to about 1 Torr in chamber

Dissipates surface charging

Eliminates the need to fully dry samples



Hazardous Samples

Biohazards (DNA, Viruses, Bacteria, etc.)

Radioisotopes

Fine dust

Toxic materials (Be metal)



Quick and Dirty Analyses

80% of what you'll ever know about something you learnin the first dirty experiment

Stabilize sample

Make it fit mechanically

Protect the instrument

Try it!


Small ParticlesCross-sectionsInsulatorsConductorsBiologicals“Untouchables”For automated analysesSemiconductor devicesManipulated samplesHigh or Low temperature processingLow vacuum observationHazardous materials“Quick-and-dirty” analysesMagnetic samples

Magnetic Sample Materials

Deflect the electron beam

High mag work very difficult

Low mag work approachable

X-ray analysis OK

Make sure pieces are stable on stage

Small particles need to be FIRMLY adhered

TEM Sample Prep for Materials

Thin Sample Prep for TEM or SEM

Dispersion of small particlesSEM: sticky dots, conductive tabs or glueTEM: alcohol dispersion on thin film

Ultramicrotomy

Mechanical thinning

Chemical thinning

Ion thinning

Image Collection, Recording and Presentation

Rule-of-thumb microscope conditions-best resolution-best depth of field-best sample preservation

Conventional Photographic Methods

Digital Methods

Presentation for:DisplayPublication

Image Collection

Proper subject identification

Proper subject orientation

Best selection of imaging conditions-HV-WD

-Spot size (aperture)-Scan rate

Subject Identification/Orientation

Representative of the whole

Image background

Not too busy

Important image information is centered and prominent

Many times a slight tilt conveys more information

“Best” Imaging Conditions

High resolution-short working distance

-small spot size-high accel. Voltage-high magnifications

Depth of field-long working distance

-low magnifications-larger spot size

Low magnification-large spot

Selection of Scan Rate for Imaging

Sensitive samples-may need to be fast

-low S/N-maybe TV integration mode

Insulating (charging) samples-decrease charging with small spot and

fast frame rate, maybe TV again-focus/stigmate in an area adjacent to the area recorded

-use image shift function to quickly move small amounts

Normally conductive samples-use slowest rate practical w/o degrading surface

Old Technology

Analog scan SEMs

2nd CRT for viewing the image as it scans

Film based camera focused on this CRT (low persistence)

Almost always a 4x5 inch Polaroid sheet film camera

Very slow scan for about a 2000 line image (~3 minutes)

P/N film or just an instant positive image

About $3/shot now

Generalized Photographic Processing

Needed for TEM image plates(Can be used for SEM film images too)

Exposure of silver halide grains (latent image)Development (reducing basic solution---> Ag0)Rinse (water) or Stop (acid)Fix (thiosulfates)Rinse (water)Dry

Scan or Print photographically

Good photographic processing results in the best imagesand are still the images that are used to compareother (newer) techniques

Newer Technology

Digital raster SEMs

Frame buffer storage of image info

Image processing

Digital image storage-usually TIFF files so that header can contain

image and microscope specific data

Fully transportable formats

Easy incorporation of images into documents


Detectors-SEI (chamber)-SEI (column)-BSE

Signal mixer-brightness-ratio

Gamma correction-corrects for desired brightness and contrast I

out~=I

in

-power function deviation from 1:11.0 darkens and enhances lower greys1.0 lightens and enhances higher greys



Gamma Corrections





Gamma corrections


Slow scan rates 1-3 continuous scan

Slow scan rates 4-8 store one frame of data-dump to disk as image file (TIFF)

Choose image pixel matrix density from 512x512 to2048x2048 (lowest is usually OK)

Right mouse button will interrupt any scan and storeresults in the buffer (incl. TV)

TV rate integration of frames can reduce random noisein the final image at a fast scan rate

File path and naming convention


Variable small raster-used to increase scan rate for image adjustment

Can store multiple images in the same frame-variable frame

-split screen-kind of gimmicky.....don't use for important images

Stereo Pair Images (Anaglyphs)

By collecting two images offset by about 4-100 in tilt

Display them side by side and cross eyes to converge

Build a blue-red image composite and use stereo glasses-In Photoimpact program:

convert images to RGBadjust color balance (red-right, blue-left)

perform image calculation (difference operator and merge)

Special Scan Modes in the LEO 982

Line scan-disable Y-axis scan to see grey-level variations

on a line

Y-modulation-if very little Z-axis information this converts it

to Y-axis deflection (not very useful)

Spot scan-mostly for x-ray data acquisition

Additional Scanning Features of the LEO 982

Dual magnification-useful for “looking around”-don't use for important images

Scan rotation-electronically rotates the raster on the sample-very useful for getting a good “presentation”

Dynamic focus-use to compensate for the portions of the sample that

fall outside the depth of field distance. Sets up aramp on the focus current +- the center of the field

Tilt correction-compensates for trapezoidal scan on highly tilted samples

Image Processing

Generally use “kernels” which are arrays of arithmeticoperators on a pixel

Standard kernels are used to blur, average, and sharpenimages. 3X3, 5x5, array of operators.

Photoshop and PhotoImpact have custom and standardkernels

Kernel Operations for Sharpening an Image

Different Kernels

Effect of Kernel Size on Operations

Contrast Enhancement

Original Average kernel

Sharpen kernel Blur kernel

Pitfalls of Image Processing

Images can be distorted and data lost

Pixelation of images

Ethical behavior dictates a minimum of processing

Always better off collecting the best image and eithernot processing or doing it only lightly

Image Manipulation

Erosion of edge pixels-kernel operator to find edges

-erode or erase edge pixels one layer at a time-break apart and separate touching features

Dilation of edge pixels-kernel operator to find edges

-dilate or add edge pixels one layer at a time-fuse separate features

Most useful in particle and other small repeating features

Presentation of Micrographs

Reports-probably least critical

-must convey information concisely

Journal-probably most critical

-size, grey-levels, resolution-must be specific and representative of the narrative

Posters-most variable in format-otherwise like journal

-conducive to point and discuss

Web-like journal

-can be interactive

Presentation Media

Photographic paper

Photo quality printer output-dye sublimation

-ink jet....getting there!-laser...maybe...

-consider viewing distance in choice

Include TIFF or JPEG files in reports using word processor

Powerpoint for talks

Micrographs as Art

Wonders of things small

Intricacies of natural samples

Subtle grey tones, like fine b/w photos

Can be psuedocolored to add interest

Comparisons to more familiar things

Explain phenomena in a “gee-whiz” way

Introduction to X-ray Microanalysis

Review of Physics of X-ray Generation

Hardware-EDS-WDS

-electron microprobe vs. SEM/EDS

Software-Spectral acquisition

-Spectral match-Qualitative analysis

-Quantitative analysis-X-ray images (maps)

-Spectral mapping-simulation of electron scattering/x-ray emission

X-ray Generation

Hardware for X-ray Microanalysis

WDS-Roland circle based Bragg-diffracting crystals and

detector arrangement-either horizontal or vertical design

EDS-cooled solid state detector-integrated FET and preamplifier

Computer accumulator/conditioner of signals

MCA output for energy vs intensity

Some hardware facility for control of the electron beamposition for mapping and DBC

WDS System

Rowland Circle in WDS Spectrometer

Typical EMPA

EDS Topics (from Notes)

Spatial Resolution

Directionality of Signals

Rough Surfaces

Hardware/Signal Processing-dead time and time constants

Microscope Parameters-overvoltage-TOA-WD (EA)

EDS Spectral Interpretation

Background Continuum

Characteristic x-rays

Excitation and absorption

Detector efficiency

Artifacts

Peak ID function (qualitative analysis)

Spectral matching

Structure of a Si(Li) Detector for X-rays

Nomogram ofE-beam Penetration

Beam Diameter vsBeam Current

Quantitative EDS Analysis

Clean spectrum

Standards vs. no-standards

K-ratio

Corrections-atomic # (Z)

-absorption (A)-fluorescence (F)

Advanced X-ray Techniques

X-ray image maps

Spectral Mapping

Particle and Phase Analysis

X-ray Image Maps

Edax Imaging and Mapping program

Process-take a look at your sample with eds

-look for elements of interest-setup ROI (region of interest) on the peaks

-start mapping function-DBC on

-dwell time-pixel density for map

-maps show up line by line in different colorsfor each ROI (element)

-color intensity is related to # of x-rays detected-can collect SE image simultaneously

Qualitative x-y spatial distribution of elements

Not very high resolution

Spectral Mapping

Sort of like previous x-ray maps

Collect full spectra at each pixel

Store data in a raw form so that it can be massaged later

Take “phases” and additively process the spectra of all thepixels that determine that phase

-leads to pretty good quantitative analysis-averages small inhomogeneities in the phase

Huge file sizes (stores greylevel and data for each pixel)->30Mbytes

Particle and Phase Analysis

Similar to mapping

Additional sizing information (area, feret diameters, calc. Volume...)

Mixes qualitative spectral matching info and morphological infoto come up with a particle or phase ID

Steers the beam on the sample to collect the data for binarized“white” areas (as determined by threshold setup)

Good for collecting statistically significant amount of data onfeature groups

Imaging Artifacts

What is an “artifact”

Sources of Artifactssample preparation

vacuum compatibilityelectron beam “issues”

too low/too high KV (not really an artifact)vibrations

stray magnetic fieldsacoustic noise

Micrograph Critique Session

Opt 307/407 Practical Scanning Electron Microscopy Considerations in any microscopy: Resolution Magnification Depth of field Secondary information.

Documents

sem slide

distance slide

nm slide

lsec slide

defocus slide

depth of field slide

gas pressure vacuum

sem system slide