Electron beam control in a scanning electron microscope----make money online

hey everyone I thought I'd talk about

some of the upgrades I've been making to

my scanning electron microscope in the

past few weeks but first I want to give

a big thanks to the FEI company for

helping me out with this new turbo

molecular pump and also a Pirani vacuum

gauge that i've been using on my

sputtering setup so this equipment is

super helpful to me and I am very

grateful so here's a quick recap of how

the scanning electron microscope column

works at the top we've got an electron

gun and this sprays out a stream of

electrons that are expanding in a fan

and the reason that the beam expands is

because the electrons are like charged

and they naturally repel each other so

even if you had a perfectly collimated

beam of electrons it would actually

spread out all by itself in this case

the electron gun itself forms a slight

lens an electron lens so the electrons

shoot off in a cone pattern and in this

particular setup the the spot at the

bottom is about you know 2 or 3

centimeters in diameter if there were no

optics in the column at all instead what

we want to do for electron microscopy is

to actually focus the beam down to a

very tight point when it hits the sample

at the bottom and so to do that we make

this column long because the longer the

column is the easier our job is in using

electron optics to focus this thing down

it's the same reason that telescopes and

microscopes are physically long it's

just because moving those photons or

electrons like take some space and what

we do is we add a lens this was called a

condenser lens and it takes the beam

that we've generated from this electron

gun and focuses it down again so there's

sort of a crossover point here just like

you'd have a crossover point with a

magnifying glass the light sort of

crosses over forms a focus and then

starts expanding again and we do that so

that we can make the most parallel beam

of electrons down here and we do this

with a combination of these electron

lenses and pin holes so you can see on

the diagram here there's a pinhole going

into this first lens and then a pinhole

down here and the whole purpose of this

whole setup here is just to get a very

collimated thin beam of electrons here

so the next step is to scan this very

thin collimated beam of electrons across

our sample so this way we can sort of

build up an image by scanning the sample

in a raster pattern basically filling up

the image pixel by pixel and we have a

couple options we could either use a

magnetic field to move those electrons

or we could use an electric field now

generally most modern electron

microscopes use a magnetic field because

you can get a stronger magnetic field

more easily so you can make the whole

column more compact and allow to do more

by by concentrating magnetic field

however the problem with that is that

you need to make iron pole pieces and

the shape is very sensitive to how the

you know the shape affects how it

operates in a very critical way so an

easier solution that isn't as quite as

high performance is to use electrostatic

deflection and it's easier because all

we need is a couple of plates just a

plate on either side of the electron

beam and then we just put a voltage

across those two plates and the electron

beam will be deflected as it goes

through the electric field formed by

those two plates and so the video today

I'm going to talk about the

amplification or the driver that I made

that drives the voltage on these

deflection plates and then after we've

deflected the beam there's a final

focusing lens that will take this

collimated you know very thin but

colonnades beam of electrons actually

focus it down to a point right on to the

surface of the thing that we want to

image I don't have the lens installed

right now but it would go right below

these deflection plates so originally I

was using a disassembled oscilloscope to

generate the voltages that I need on

these deflection plates to move the

electron beam around and this worked out

pretty well because the oscilloscope

already has amplifiers to amplify a

small input signal could only be like a

volt or even less and amplify that all

the way up to the hundreds of volts that

we need to move the electron beam around

and conveniently oscilloscopes already

have an electrostatic cathode ray tube

that's that's how they say old analog

scopes worked and so all I had to do is

unhook the deflection plates in the CRT

and hook the scope circuit up to the

deflection plates in my electron

microscope so it's sort of a really low

work pretty high performance solution

however I did end up running into

problems the voltage available was only

about I think 150 or 200 volts which

wasn't quite enough so I couldn't scan a

very big area because the voltages were

just not high enough to move that

electron beam enough and the offsets or

the available range of offsets was

limited because of this and that was the

sort of the biggest problem so I wanted

to have a circuit that I would be a

little bit more adaptable to what I was

doing and also mainly just use a higher

voltage and I found this circuit made by

rhe head Wahby

I hope in getting your name correct at

web de fête org and this guy builds

Silla scope clocks and similar things

and also needed a deflection circuit to

drive CRTs so I've modified his circuit

a little bit but the main gist of it is

that it's a single ended to differential

amplifier so we put our positive voltage

across the top here and then we have

common emitter single transistor

amplifiers and they're coupled together

with this pot here so as we put us an

input signal here let's say is 0 to 5

volts these amplifiers this was

non-inverting and this is inverting and

so what we have up here

are two signals that are out of phase

with each other and that's good because

that means the average voltage between

these two output lines is always going

to be about the same so that as one goes

up the other goes down and vice-versa

but they always average out to about the

same if they didn't if there was a

common mode change where if the average

voltage did actually change then these

deflection plates would start working

like a lens but unfortunately they only

exist the deflection plates are only in

one axis and so what we'd have is a lens

that only functions in one axis and not

the other and that would cause a

stigmatism which is basically with the

beam becomes sort of oval shaped instead

of round now if you can control that

that's actually a good thing because we

can correct for a stigmatism that's

caused

by other things in the system it can

actually make it astigmatism corrector

however if we don't have control over it

then you might get a stigmatism and you

know you can't correct for it this

circuit will introduce a slight amount

of common mode voltage changing and you

can't actually control it that easily

but I'm working on some tweaks that will

fix that the transistors are ztx 458

which are a good high voltage transistor

that can take up to about 400 volts and

also have pretty good gain I think the

gain is about 200 on these and so I've

set the circuit up with a 350 volt rail

here and I'm using 470 K ohm pull-up

resistors here so it's a basic common

emitter amplifier where if we put some

current into the transistor it will pull

down on this resistor here and lower the

voltage here so when the transistors are

not conducting anything the voltage on

these lines is very close to the rail

and when the transistor is fully

conducting then the voltage is very

close to ground although it's a little

bit off because of the emitter resistor

there originally the circuit had an

additional transistor in here so that he

could use a much higher supply rail

since I don't have that set up on my

circuit here I just took out that

transistor and then lowered the supply

voltage down to about 350 volts this pot

controls the offset basically where the

scanning is going to happen it basically

just moves the whole center point of

this amplifier and this pot controls the

gain so if this is a low value then the

coupling between these two amplifiers is

very tight and you'll get a very high

gain so I soldered the circuit together

on some vector board and added an op-amp

just to provide like a voltage follower

on the input so the input impedance to

this amplifier is relatively low and I

wanted to drive it from a PWM output

which is going to have a relatively high

output impedance so the op-amp fixes

that problem I'm using an old rackmount

high voltage supply for the 350 volt

rail and the current draw in sort of a

worst-case scenario would be

two or three milliamps okay let's fire

it up the first step is to turn on the

roughing vacuum pump and you'll actually

be able to see the blades of the turbo

pump turning just from the air being

drawn through it

okay so after about five or ten minutes

the thermocouple gauge is showing

between 500 and a thousand

militar and I'm going to switch on the

turbo pump and the readout here is a

zero to ten scale that indicates the

speed of the turbo pump so this this

pump has a built-in controller and you

basically just tell it to go on and it

will ramp up the speed automatically so

when the reading is 10 10 volts on that

meter the speed is about 70,000 rpm and

as you can see look how fast the gauge

declines the turbo pump is able to start

pumping from the 500 militar range and

take it all the way down to high vacuum

in sort of one swoop

okay so we've been pumping for about 10

or 20 minutes and we're down to about 5

times 10 to the -5 millibar so now we'll

move over to the control side and turn

on the filament and then also turn on

the high voltage supply so I've got a

bit of phosphor coated glass in there

taken from an oscilloscope CRT and we

can see that when the power comes on

there's now a tiny little green spot

where the electron beam is hitting the

phosphor so the beam has been expanded

and collimated by that series of lenses

and pinholes that I talked about but

there's currently no focus and no

deflection happening however if I turn

on the deflection amplifier we can see a

little Christmas tree so I borrowed this

code from John DeCristofaro known as

John Jr who came up with this cool

little hack a few years ago and he's

using an Arduino to generate PWM and

then the PWM is smoothed out into analog

voltages so you can feed this into your

XY mode on the oscilloscope and have a

nice little Christmas tree graphic there

so I can use the control pots in the

amplifier to move this image around and

you can see that when we get close to

one of the extremes it will start

bunching up because the deflection

plates have sort of run out of the

ability to actually move the electron

beam but as you can see the range is

actually pretty good and of course we

also have control over the scale

here's a more typical raster scan

pattern so this is covering a square

area and it's being scanned very slowly

so you can see the movement of the beam

here's a back and forth raster scan with

digital acquisition it doesn't make

sense to return the beam all the way to

one side and so here we're just scanning

back and forth and covering a whole

rectangle that way and here we are going

much faster so now you can see the area

that it's actually tracing out and of

course we could do the back and forth in

the y-direction as well instead of

returning to one edge like it is now

unfortunately the pedo boom out but from

the Arduino is not even close to fast

enough to get a real-time image out of

this least I don't think and so I think

what I'm going to have to do is

implement the microcontroller with a

true DAC to get the bandwidth into

accuracy that I want out of this but

that'll be a topic for a future video ok

see you next time

byeearn money online without investment for students, earn money by clicking ads, earn money online without investment, online earn money website, online jobs to earn money, best online income site, top 10 online money earning sites, easy income online, easy online earning, earn money online from home, make money online legit, earn money online free fast and easy, online earning websites list, genuine online money earning sites, online work to earn money, online surveys to earn money, earn money through internet, best online income, earn money online data entry, easy ways to make money online, best online earning websites, top websites to earn money, online typing jobs for students to earn money, earn skrill money online, earn skrill money, best way to earn money from home, make instant money online absolutely free, trusted online money making sites, online income site, best online earning, money online, earn money from home, earn dollars online, earn money online, earn money online 2019, earn money online by typing pages, earn money online daily, online work at home and earn money, online earning,