Probe formation and basic image formation in SEM

Probe formation and basic image formation in SEM Using JEOL 5310 LV as example http: //virtual. itg. uiuc. edu/download s/esem-manual. swf

Electrical currents in SEM column (those in red are or were part of the current that comes ‘boiling’ from the filament): CURRENTS filament heating current to heat filament and cause ‘boiling’ of e- off of tip (m. Amps-Amps AC) load current or emission current; gun current that doesn’t make it through the hole in the anode plate ‘lost’ at anode (μAmps-m. Amps) beam current; current in beam as it traverses column (p. Amps-m. Amps) probe current; current in beam as it hits sample (p. Amp-μAmps) sample current; current that leaves sample and enters SEM beam circuit through the column ‘ground’ (probe current minus secondary and backscattered e-) lens currents; current flowing through condenser and objective electron lenses and the scanning and stigmation coils to produce magnetic fields that steer or focus the e- beam (m. Amps) Controls to effect and measure these currents: filament heating knob, gun bias, ACCV, aperture size, condenser lens, objective lens, contrast, magnification, stigmator control, faraday cup with ammeter

Voltages in SEM column : VOLTAGES acceleration voltage (ACCV); this is the voltage that defines the energy or velocity of the beam electrons and the energy of specimen emissions (usually 1000 V-30000 V, anode and column are positive relative to filament/cap). Gun bias voltage (all voltage is really a ‘bias’); voltage between filament and Wenhelt (gun) cap that shapes the e- beam at its origin and keeps the filament current in feedback at saturation (hundreds of V, cap is negative relative to filament). Faraday cage voltage; voltage relative to column or ‘ground’ that either draws, repels, or neither the secondary e- to/from the ET detector (-100 V - +300 V). The cage also prevents the field caused by the corona ring/scintillator face from distorting the beam. Scintillator face voltage; voltage relative to ground, metal coating on scintillator accelerates low energy secondary e- to induce light signal generation at the ET detector’s scintillator (‘phosphor’) surface (~+10000 V) PMT (photo multiplier tube) voltage relative to photocathode target at front of PMT; this is a gain voltage that multiplies the signal at the ET detector. For example a gain of 10 may give 10 counts out for each 1 ein or 100 counts out for each 10 e- in. (+100 V - +1500 V). Control this V with contrast knob. Controls: ACCV, gun bias, backscatter detector active, contrast control

The self-biased thermionic tungsten (W) electron gun (AC) Electric field lines result from an electrostatic field formed by the voltages maintained between filament, cap and anode. Only when positive field interacts with filament will e- leave the tip and travel through the hole in the cap. (red arrows indicates direction of e- movement) One way to look at the self-biasing gun (remember, without electrostatic fields, e- boil from hot filament in all directions): 1 -all e- that leave the tip must be replenished by ‘traveling’ through the circuit (through the anode and stage to the high voltage tank to the filament) 2 -if there were no bias resistor the cap would not control e- emission, the filament would emit at the maximum current allowed by the HV circuit, e- and tungsten atoms would boil off the filament until it broke. 3 -with bias resistor, as filament is depleted of e- it becomes relatively more +++ to the cap. The field present is able to draw fewer e- from the filament which slows e- output. This results in filament becoming relatively less +++ thus we see a temporary increases in e- output. This increase returns the filament to more +++ relative to the cap……………. . ALL ELSE BEING EQUAL, AT SATURATION, FILAMENT OUTPUT IS A FUNCTION OF BIAS RESISTOR SETTING (gun bias)

emission from filament at e- gun W filament

condenser lens current constant, objective lens current changes objective lens current constant, condenser lens current changes alpha is convergence angle P is object distance (lens terminology) q is image distance (lens terminology) in this diagram q is the same as WD (working distance) d is spot diameter f is focal distance (variable; higher lens strength, shorter f, greater demagnification of spot, smaller spot) All beam current shown by the hatched lines is lost from the probe as it enters the column walls or apertures, the smaller the spot size, the more beam current is ‘lost’.

In the e- lens, current through the windings creates a magnetic field, the pole piece is an important part of the lens design; it concentrates the field at the region where the focusing action is desired. For your information only; the fields that form the e- beam in the e- gun are electrostatic fields (also known as electric fields). The fields pictured on this slide (from current carrying coils) are magnetic fields. Ernst Ruska, credited with invention of the e- microscope, chose magnetic fields for his lenses due to practical, engineering concerns. There actually 2 pole pieces in an e- lens.

Just for your information from SPI. • • SPI Tungsten Hairpin Brand Filaments and Electron Sources Some easy to understand comments about tungsten hairpin vs. La. B 6 sources for electron microscopes This is not intended to be a detailed description of how the electron sources in a microscope operate, but to outline the salient details. Detailed descriptions abound in the literature and various EM reference books (see for example. Reimer - Scanning Electron Microscopy Springer- Verlag 1985) The first thing to make clear is that the accelerating voltage of the microscope (HV) does not have a great deal to do with electron emission per se. There is an interaction of the filament, it's power supplies and the HV power supplies. This is due to the fact the filament and it's power supply must reside within the HV tank and operate while electrically floating at the high end of the HV terminal, also it is relevant to design the electronics to simplify operation and hence there is some interconnections between the HV and the bias voltages. The important point to keep in mind here is that the primary function of the HV is to accelerate the electrons into the electron optical column. The HV itself is not the source of the electron beam. It is the function of the filament to be the source of these electrons. There are 3 common types of electron sources in use today: Thermionic emission, Partial Field Emitters, and Pure Field Emitters. In the case of a thermal emitter, electrons are emitted from the hot material in accord with the Richardson equation for thermionic (heated) emission. The function of the filament (heater) current is to give the electrons in the source enough thermal energy to over come the energy barrier which is prohibiting their escape. All metals will give off electrons when heated and the higher the temperature the more electrons are emitted, however most materials do not survive very long at the temperatures required for this process to yield a significant amount of electrons. Tungsten (W) has a sufficiently high melting temperature (3650 K) that it survives for a longer time at elevated temperatures ( ~ 2600 - 3000 K) before it fails an it is the material of choice for a typical thermionic emitter. Another way to get emission is to decrease the "work function (energy barrier)" which is limiting emission. Materials with lower work functions emit at lower temperatures (La. B 6 for example). These materials are sometime referred to as partial field emitters since they exploit the fact that their reduced work function enables their operation at lower temperatures than W (~1400 -2000 K). The use of this "partial field emitter" nomenclature is not rigorously true, since the primary excitation mechanism is still thermal activation. Finally field emission sources operate due to a decrease of their energy barrier width by an applied field permitting quantum tunneling of electrons through the narrowed barrier into "vacuum". This is a different process all together than thermionic emission and is governed by the Fowler- Nordheim relations. There are two types of Field emitters - Thermally assisted (sometimes called Schottky) emitters and Cold Field Emitters. As the names imply, thermally assisted Field emitters are heated, while the Cold Field emitters are not, both require the application of an external field to "extract" the electron beam from a very small tip. The electrons which are emitted from the filament regardless of the type of emitter are "replaced" by virtue of the fact that there is a connection to ground and the electron reservoir in metals is a continuum, hence the filament remains overall neutral. If this ground connection was not present, the material would become in effect, locally charged, and the electron emission would start to decline, and eventually stop (the local energy barrier would continually increase). For the case of the Thermionic & La. B 6 Guns, the filament, wehnelt cap and anode from a triode system. By judicious design of the power supply the wehnelt is made slightly negative with the filament and the resulting the electrostatic field which acts as a small lens and produces a focussing of the electrons which are emitted only from a limited region of the hot tip of the W source. This voltage difference is usually called the "bias" voltage of the gun. In the Field emitters a multiple anode system is employed instead of the triode configuration. Here one applies an extraction voltage instead of the bias voltage. Bias voltages are usually a few hundred volts, while the extraction voltage on a FEG is usually in the range of 3 -5 k. V. The filament and wehnelt (or last anode for an FEG) are also connected to the negative terminal of the HV supply. Thus, the beam of electrons which leaves the wehnelt is accelerated by the HV toward the nearest grounded anode, which is judiciously placed at the base of the gun. This acceleration gives the electrons their Z (axial) velocity. Some of these electrons make it through a small aperture in the grounded anode which is the entrance to the electron column. Once past the final gun electrode (ie. the one at ground potential ) the Z acceleration of the electrons is essentially zero and they are traveling at constant velocity (determined by the HV) until they hit the sample. The function of the remaining lenses and deflectors in the body of the microscope are to further deflect/focus this beam onto the specimen in the manner chosen by the analyst. Lastly you should not that the electrons are emitted from a thermionic filament much the same way as they are in a light bulb. As in the case of the light bulb, failure of the filament is usually due to mass transfer/melting. This does involve movement of metal atoms away from the point of failure (even though you are below the melting point). Over heating the filament causes this to happen faster and thus decreased the filament life. The above information was taken from an E-mail reply drafted by Dr. Nestor Zaluzec to answer someone's question and we re-publish his E-mail comments here with minor editing for greater clarity. We wish to thank Dr. Zaluzec, Argonne National Laboratories for permitting us to use this short description of the differences in the workings of the different electron sources.

How does energy travel through the SEM from gun to image? electrons - light - electrons - light beam - sample - scintillator - PM gain - other amps. - CRTbeam - CRT screen C = S 2 -S 1 / S 2 C% = (S 2 -S 1 / S 2 ) x 100 C is contrast of signal at sample S 2 & S 1 are signal at 2 points of interest at the sample where S 2>S 1 light Different types of signal are generated by sample/beam interaction and can be mapped by the SEM (MU does not have an Auger e- detector). This diagram of an ET detector (secondary edetector) is only one of the detectors that can map signals from the sample/beam interaction. What other detectors can we use on the JEOL 5310 LV? Adapted from Goldstein 2003 In the graph at right, each line represents a beam current that is capable of generating an image. Beam current correlates with spot size (high beam current=large probe diameter). This graph shows how inherently low sample contrast can be overcome by slowing scan speed or raising beam current. What is a cathode ray (as in CRT)?

e- beam of constant ACCV Si Co Si signal Co signal (Se- per location or pixel) Absolute difference (Co minus Si)in # of Se- that can be multiplied by the ET PMT probe current = 1 n. A or 6. 25 x 109 e- / second fast scan dwell time = 0. 1 second slow scan dwell time = 1 second 5 x 109 5 x 1010 6. 25 x 109 6. 25 x 1010 1. 25 x 109 1. 25 x 1010 probe current = 1 u. A or 6. 25 x 1012 e- / second fast scan dwell time = 0. 1 second slow scan dwell time = 1 second 5 x 1012 5 x 1013 6. 25 x 1012 6. 25 x 1013 1. 25 x 1012 1. 25 x 1013 SEM variables that allow us to increase signal differences given an inherent sample contrast (probe current and dwell time) Inherent sample contrast (these are convenient numbers, not true values for Si and Co) These are 2 sample locations, location 1 is Si, location 2 is Co. Lets say that when a single beam e- hits Si it generates 8 Se-, when it hits Co, it generates 10 Se-. What is the inherent contrast in terms of Se- of these 2 sample locations. C = S 2 -S 1 / S 2 C% = (S 2 -S 1 / S 2 ) x 100 CSi/Co = 10 -8 / 10 =. 2 or 20% IN WORDS ONE COULD SAY, “GIVEN SAMPLE CONTRAST OF 1% AND BEAM CURRENT OF 1 n. A, WE NEED A FRAME TIME OF AT LEAST 600 SECONDS TO DIFFERENTIATE 2 SAMPLE LOCATIONS. ” or , “GIVEN SAMPLE CONTRAST OF 1% AND FRAME TIME OF 100 SECONDS, WE NEED A BEAM CURRENT OF ~10 n. A OR MORE, TO DIFFERENTIATE 2 SAMPLE LOCATIONS. ” dwell time = frame time / pixel # = # of sample locations being probed and displayed

The Everhart-Thornley SE detector SE-secondary e. PM-photomultiplier LG-light guide S-scintillator F-Faraday cage B-backscattered e- +300 V (relative to ground) SE -50 V (relative to ground) Adapted from Goldstein