Sem PDF
Sem PDF
Sem PDF
1 % 23
or 111:igncuc ficM~ '111u, dungc I\ '- ommuum de pcndmg on the clccu 1c. and nug11rtJC field E k « ~
1cmcs, "Inch ha\'c Iewer .lhcr riuons than c1cc.trmuuc lc11SC\, .ire used m electron m1croscopy
(a)
(b)
Fig. 2.1 : (a) Intensity distribution at the image plane forming an Airy di$k The point5 are considered resoh-ed
ii the intensity maxima are '>eparated at half the height. {b) Rayleigh criterion of resolutJon.
variable rnamen
t power supply
Filament
Wehnelt
cylinder Bias
(Grid cap) resistor
High
wltage
~
+
~~
\ cunertl.
Beam current
Anode Plate
\\ hc-11 tht' st~k llll.td a, CnLa,t',! m ,t ca,;c ,,f,Pll mm, tlw 11 1,1gnl'llC tidd ,dong I he.· ,1.xi-; is inrrc,i~cd If
the
:mnu I.u
!,!a}' 1, 111,hu.· nl ,nit 11011 ••1 gn:,ttc.·r ro11cc1Jtr,1tio11 of the Ill
•
J
1
~-
~ , ,"ICU <"Xl'X'pt .at a n.in'l1\\
1 • •
;ignetic
ltr'd
,'t
-'·
Iii.fl.'\.:~ "-
~nort
..
a:\."l.lll d1<.ta1h~ ou ur,.1 he .
,,,,1,·ra-c,·,
.u,· tht·,t• .111nul.1r rinh" which focus the 111.1gnct1c
. .
.
• _..., • typi cal imtrum cnt~.
lbou at .a ~,~n IOC'..1t1on.111~ held ,tn'u~>th u,,·d 1, bd,m· 20.()()() g.tus\ 111
ObJect
f q
- -- - ~ - - Image
(a) (b)
tically correct ed
The foal Jengrh is given by f = Ki, (., '1)2 w here K 1s a constan t; v, the relativis
aooele nnng ,'Olcige; and /\/1, che numbe r of rurns of rhe coils. Focusin g is achieve d
by varying the current
to current . The foc.tl
throug h che coils. le is impon.: rnr ro nore char rhe foca l len1:,rrh 1s not Jiuearly related
is affected by
Jeng-Ji of die lens is directl r propor tional to che accelerating voltage . So the image quality
\."2I)'JJlg me electro n \'eJocit}~
there is no w.iy w
Lnu cbtn:m.ons: Electro n optics suffers from abc rr.ttions . ilur unlike in light optics,
sohc diese :aberr.m ons.AIJ chat can be done is ro rt.>d uc:e them through proper
design conside rations.
more strongly. As •1
SphnicaJ aba,-atJ01t: Eleccrons 111 far away tr:ijecrnne:. frnm tlw optic axis are bent
different spot
rc:suh. the electron beam emenn g the lens ucar d1c edge of the lt·ns, is brough t to focus at a
~ the SJ>OtS closer to the centre. ~l his is schc111 ,1t iL·ally rnustnated in Fig. 2.4(a). The error in the image
axis of the lens.
u.e to this become s rnore pronou nced as the uc:1111 is 111ovcd furth t'r .1way from the optical
Im es1 g t n and Man pulat ng Mat rials m the Nanosu 27
DiffC":-rnu21 focmm!? ousc, the unage at the penme,er to £f, smeued ~ d of 1: the ttntre As a r •
0
~ ~ ; lb: mu~ 3 Pt'('.:l~ 25 3 dBk ~ld not as :a pom: The ~mall~t disl 1s called .he sphenal abttra·
di i-: 0 • le~t con,''\Non. d, = 1/2 Ca \\ here C 1s the sphenol aberr.ioon coeffiaen: and a IS the angle
of the ourer m cluou;h the leru. C, 1.5 typically :a tew mm for lm.-e5 with shor. foal lm.:-.hs The sphenul
abcrnuon cw be nururruzed by removmg the ou.er edce of die beam. This 15 acruC\·ed by placin£ a small-
holed aperture ,lt che centre of .he nuc;ne~c field or ~'lledu..e}y belaw 1,. Ho\\C\-er. :a muller du.me·e~
miuc~ the beam cur-rem 211d ilia leads to aperture dtffi.tcnon.
OrrMWiU abm.;w.n;: This abernnon is due to the eneq~·y spread of the electrons. \l/hen li~ of c:li:E"r.-e,r
energi~ eme~ a converging lens ;u the ~e po1m, che extem of ddlec.:ion will depend on ~ie rner-g-,. ln
light opcio. r.diarion ot a shorter wavdeD.£Cb 15 deflected more strongly .h.an ~ of :a lon,;er '\"\'J\~lrng-.h.
In electron opacs. che re\"erse happens. 1.e. a ~honer ,u,-eleng-..h is ddiec...ed l~ woru::h·.This IS due .o .he
t"".h.-c .hat elecrroru .ue subjected to les.ser dedecoon when th~ beam energy is hi£h.. ~:s·.
resuli. die be.ams
of ~"O diE"erem energies form inuges .u cliferenc poincs as illuscra~ m Fig. 2 ..!(t>). Due ,o chis 2ber.-a::ion.
Ul.Stead oi , point. a dtsk results and che du.merer of ilie disk o: le~c confimon cm lbe ;i,-en as.
J = C a (1 EI E.). where C IS the chromic aberruion coefiiciem. E. is t h e ~ ene!Ff :and !1£ is :he
enerm spread. The fracriorul ,-ariar:ion in beam energy IS the significant iaaor.
Chronucic aberration cm be reduced by subilizing the energy 0£ :he elecrron bewi. Sr.wilizcd
accelennon \"Oh:age and impIU\-ed gun design enstL"'C che s:.lbi!i..-y oi ciie i:>eam.. 11'le e::ec oi ch..-ommc
~rracion i.s pronounced near the perimeter oia converging lens and an aperrure an be used w dimim:e
these elearons.
Apmurt d,ffraaion: The wave nature of elearons 0.1..1.ses the beam to diffi::act upon pming through a
rurrow slir. &ch beam passing through che slit Sets up ics own ·waves. These will incena: co gn-e 2 bright
spot in che middle and a ser of concencric ~ in che inuge plane called the ·Air)· disk·. If the intensm
discnbution is plotted in one dimension. ic looks like due one sho-..m in Fig. 2.4 c). The concn'buaon ro
the spoc size due co dilli-acrion is given as die half dimiecer oi die Airy~ d1 = 0.61 }. / a , where A is
che waveleng--..h of che beam and a LS che aperrure half angle. In order co reduce rhe dfea:s of dttfurnon.
it is necessary t0 have as greac an angle as ~1ble be.ween che opnol 1-'0S md the lens pmmner. Tlut
would amoum to having no aperrun: ar all. Bu: a smaller apenure is needed lO reduce the dfecn of
sphencal aberration and chromaric aberration. wtuch would cause diifracrion problems.Thus m oprunum
aperture size must be chosen.
AJcigmacism: Asrig:nmism refen t0 the; improper sh.ipc oi che beam.:\ pomt obJeet is focused co two-
line foci at the IJ'i'Uge pbne and UlStead of a point. m elli~ appem. The rwo-line foa DU} be fort't'd ro
coincide for correcting this defecc. A.srigrrurum occun due co defera m die foamng lidds. which could
be due to severa.l ~peccs reb.ted co elc:ctromagnecic lenses, aperrurt1 and omer column components.
Imperfection in nuchining an ause 21ng.numm. Asrignumm B cometffi by :.a set of nugnm ailed
'srigmawrs', which are placed a.round che c1rcumfrrencc of d1c column These att adjusted :.acronhng to
mength and p01ition in an effort co mduce an equal and opposite etfeCl on the ~
The effect of lens aberrations is import.mt for the obJtCO\"'t lens :n the dferu aUSttl to the be2m
would be mull in comparison to the diameter at ocher len.ses.Typ1o.J sphencal :.abm.rion can be corrected
~o et poi nt p Object point
. , '
, , '' '
..
, ,
,, ''
''
''
,, , ''
, , ''
, ,, '\
A
,,
a ,,
• Bea m at the
per iph ery 1s
bent mo re
Imago Ima ge
pla ne pla ne
Sph enc al Chromatic
abe rrat ion aberration
(a) (b)
Ob jec t point
Ob jec t
Len s
len s -all~-
Disk of least
Image
pla ne ......._,.,_,......,. confustion
/'-- - Diffraction
pat ter n Line focus 2
(c)
{d)
rration,
gn efl c lenses. (a) Sp her ica l aberration, (b) chr om ic abe
of e/e rtro ma
Fig. 2.4 : Various ah crr ati om a'iliRmatism. Ad ap ted from Golds
tein, et al. 20 03 (Ref. 1).
re ddfr,1( tio n, ,m d (d)
(c) ap cnu
and Y axes (Fig. 2.5).T he scan coils lie within the column and
move the electron beam as per the requirement
across the specimen. They are electromagnetic coils and
are energized by the scan generator. The scan
generator is conne cted to other comp onent s such as the
cathode ray tube (CRT ) and the magnification
module.
Th~ scan is made as. follows. The electron beam swept
15 across the sample. The patte rn over the
sample lS synchronous with that observed in the CRT. The
secondary electrons produ ced by the sample
are detected. The intensity of the signal at the CRT is propo
rtional to the secondary electrons. An intense
signal can illuminate several dots on the screen, while
a weak signal would mean that no dots will be
illuminated by the electron gun. The detec tor therefore
gives the intensity of the signal, wlule the raster
pattern gives the location of the signal. In this way, the
image on the CRT is built up point by point to
match what is happe ning on the surface of the sample.
This mann er in whic h an image is formed is the essent
ial difference between the transmission and
scanrung types of microscopes. A couple of impo rtant obser
vations need to be made in this type of image
formation. Firstly, the focus is depen dent upon the size of
the electron beam spot. The smaller the spot on
the sample, the bette r is the focus. Secondly, magnification
is not produced by a magnification or enlar gmg
lens but rather by taking advantage of the differential betwe
en the size of the scan pattern on the sample
and the size of the CRT.
The size of the CRT is fixed.The size of the scan pattern
on the sample is variable and is deter mme d
by t~ magnification module. By narrowing the size of
the area which is scanned and conveying that to
the CRT , we can increase the magnification of the image
. The smaller the area scanned, the lesser is the
distance betwe en the raster points, and the smaller 1s the amou
nt of current needed to shift the beam from
point to point.The greater the area scanned, the lower 1s
the magnification, while the greater the distance
between the raster points, the greater is the amou nt of
current needed to shift the beam from point to
point . In this way, when we operate the SEM at relatively
low magnifications, we actually push the scan
coils to their extremes.
The scan gener ator changes the step curre nt to the scan
coils. This curre nt is then multiplied by a
constant by the magnification modu le and sent to the scan
coils. The higher the total magnification, the
lower is the multiplier constant.
Sea coils
~ Sea:, gen!."lt!On •
~ t --- --~
I
L
-!/JV to •250V
Fig. 2.i : Smem.J!,c OI an 5£\l The electron beam r, sc,anned b) a ~11015Cm ro,l, andthes«ondMyelec:tron5
t dereeted b) the cktecr0t. 8\ apph ng a negative potenllal to the Faraddt c• ~ secvndM)
,:-J.,.-rm,,.,..· ,a t: ,eyeaed comp!ert l. Adapted trom Goldstein
et ,1/. 2003 (Re,: IJ.
e pru'tla l opn.uo n ofthe.- dc.-tl"('tor, :a chm m(U) c~nng as applxd on me u6ce al'dat•
•Milac•-:
,-c potc.-ntul ts :applied to the.- mml surface so that all d«trons. induding die b, ~
not 6t tbt
tttro an- accdemcd to 1t so as to generate photons ~ high wkage should
.and fj r dm rc~son. a faradJ\ age 1s kept over the santilbcor, which as aecaicabaulalrd.
• desired potfflw1111 th.: range of -50 co +250 V to the Faraday CJF, a con.-. 111 m,n or
11,t\ l'll 1 ,t,,l "h11 h ,, a h" drn11it ll i, ,rnd du <let ,nm,< 0111111 i 0 111 11 th oth r 11d u
l 111 111
I \ I l\)'hh ,1 I\, llH 'l \ Olllbt \ cit I I Mil ) ,rhl llll I( I \ ( ' Ill tlH low ) (, V ,11,wc, lw111 I pl It t ork
"' h,11,\\ 11111~ 0 1i11.i11011 l u" 111r1n) 'liimd u ) dttllrnl, H;q 11 m: 1ctclc1 111 n11 lor cl C Cllfl byJ
,!111 nd ,,1n11
Secondary electrons
ContmUOlis \ -•avs
Sample
Mi<nwaalysis l resoluti
io n nu cro an alv s1s of ma cenals is possible. Th e spatia
hig h spa cia l resol ut citat:Ion beam, wh ich is of
the order ofi
fn bo th SE.i\1 :m d TEJ\1 . all dim en sio n of the ex
,:-.i b]e by rh e <:m us excitations in
of cb e an aly sis is nl 3d e po ms tru me nt s. T he ele ctr on be am causes vario
fev, na no m et er s in .-.t3 te-
ot:.rhe- ar t
sen t in the ma ter ial . Ch aracteristic X-rays emitted
pre
lcrer i,cfr of th e ele me nts identification. Th e int en sit
y of th
su np le. wh ic h :u-e c.har. be us ed fo r ele me nta l
co re ho le de ca v ca n
th e sa mp le as a resuJ r of
cit3ci\-e an ah -m . y dispemv
si,gna.1 an be us ed fo r q u:m rre sp on ds to the en erg y analysis called 'energ
m cwo ways. On e co elength dispersn:
f\1 ..ia oa iul ys :is is do ne .sp on ds to wa ve length analysis called 'wav
ile th e oc he r co rre em
~ m er ry " (E DS ). wh uc ion 1s po mb le in W DS , it is mo re cumbersom
hi le im pr o\·ed en er gy re ,ol or tio na l to bo th th e energ
y JIJ
sp ea ro rn ea y• (U 'D S) . \~' m the de tec to r 1s pr op
S. In ED S. J si~'llal fro ray are de ter mi ne d separa
tely.
am e- co ns um m g du n ED en gth an d in ten si ty of the X-
me X-'n,·s. In ~ ' DS, rh e wa \·d
- Th is results ·
mt cn sm ., of
. m. it ge e, de ce ler ate d du e to the co ulo mb ic field. 0
ch es rh e aro to as bremsstrahlung
U' :he n 2n de co -o n ap pr oa erm · ap pe ars as ph ot on , ref err ed
ch e ele ctr on an d rhac en
1
.a Joss of en er gy fo r all en erg ies rill the en erg y of th e or igi na l dectr0'.
s of
l'.2d iario n con rains ph ot on electrons. Th e characteristic
X
'br ea ki ng nd iat ion -_ Th is en eq,> '}· of the pr im ary
as an d. ea ro n aa n los e ~m
y en ergy, fro m ze ro to rhe thl y \~ Jry ing ph ot on intensity. Th e«
J
s on chi s far ge . sm oo
D) 'i em .m ed by th e
ato ms will ap pear as sp ike n be ide nti fied. Th e intensities of the
se Jin~ cJ
an aro m c.r
\• lin es wi th wh ich tho ug h various pa ram ete rs
deterru in
se ve n) du r:u u: ris oc X- f3 g .. pe cie !> in rhe sam ple .
be related to d1e concen
tr.acions o f di e em itt in ion int t·nsity are, in ne r shell ion
iz.ation Cf\)),'
ace d ro X - ray ern hs
ru m asp ects ref
the imcnsmcs. Th e im po o f ma terials an d X- ray pr od uc
tio n range.
tio n cro ss se cD on
s.cction, X-ray ab so rp
fnH hgattng :ind Ma111p11lc11ing \\atcrial~ In the N,mosrnlc
1 Elh a pllonm, emnted l" the sample a, r collcucd ,111d mc.m,rccl su1111ltanco11~ly
hy a §Ohcl ~Ute
K'42\ &tectu~ llu~ oommon EDS dc1r~tot ,, l11h1t1m d11ltnf ,,Ju 011. S1(I 1). l111r1m1c or high punty Ge
iPGc u al~ used, wfuch ~ nrnrc l o1n111111111t tlw <~~col r I M umnmH·11t~ due to then higher (;ullccuon
dnOO'K"\ \Vhen an X-n, photon fall, on ~11 11Hr m,ir \C1111rnndtH tnr (h.1vi11g no (;h,1rgc c~rr1cn), due to
ph ~puon, durge camcn (clcl lJ\111, .md hole,) arc crc.1tcd. 1 hc\C .ire \Wept by ,111 :1pplred bi:u
a du.--ge pul,;c illu, charge pul,c i, then rnnwrccd imn .t volt.1gc qb'fl,11. lmran~ic con<lmort 1s hird
w aduC'\'t' md det~ctor c.rpuh arc m.idc lo bch.1w hkc intnn,ic ,,)icon. J'hi, i, m.1dc hy Jpply111g [ 1 on
p-cype ~Ulet:bv fomun~ .1 r -nJUllltion.Thejunct1on region will behave ltkc m intrin,ic \ertuc.:onductor.
The " ~ ot dmrregion c.m be e,~.mdcd by applying :an electric field at elevated temper.icures. The
l'CD0\':21 o, most o, the p-tyt"e ~1 re~1on m.tl.:e\ a detector (which i\ the 51 left behind). l lowcver, the Li
present u mobile at room temperature and it is ncce,\arv to cool down the detector 10 77 K in order to
ope.r.te the detector under bi~. '
The w:rmsic ~con acove layer i) covered on the front 5idc wnh /Hype S1 ,rnd in the b.1ck with 11-
npc Si.The front 15 :abo coated ";ith a tlun gold layer for electncal concact. The device is kept behind .1 Be
wmd°'" aiuch bloc.1.s visible light. Also, it mechanically seals the detector assembl~ from the vacuum
dumber. thereby :n-01ding contmunaaon. When an energetic photon strikes the detector, ch.irge earners
~ OC2ted. The detector is rever~e-b1ased, 1.e. p--type 1s connected co negaave potential. This means that
,._,1icn the durge carriers are created. the holes move to the p-sidc while the electrons mow co the ti-side,
thus ~ g a pulse of eleco-ons at the 11-s1de, which can be amplified. Jn order co reduce noi,e and also
Li mobilit)~ co m.a.imair1 the intrinsic condmon, the detector 1s cooled. When photons of high energy
2.bo-."C the S1 K sheU binding energy (1.84 keV) fall on 1t, phocoeleccron emission c.1ke, plJce. The
pbcr..ode.co-on loses enerb'Y inelastically, thus excmng electron-hole pairs.The core hole produced may be
fi!kii b-y phoroenussion or Auger emission.The photon may be reabsorbed resulang m phoroemb~ion, or
11 m.zy undergo ine!a!ric scattering.Thus, ail che energy of the initial photon is used 111 gen er-a ting electrons.
In fn'OUnbJe conditions, che number of charges created in the detector per photon i, equ.tl co photon
energy/energy required for che creanon of an electron-hole pair (e) For a 5000 eV photon, chi, works
out to 5000/3.86 or 1295 electrons (e ac 77 K is 3.86 eV for Si). However, chis number i, wry ,mall.and
gener.ates a \ery small charge:: pulse of 2 x l 0 H, C. Each charge puhe, convened ro voltage, repn·~enb the
ent"rgy of the- photous. The resolution of the photon energy, d1stingui,hablc by solid ,c;lte detc:-ctor, of this
i,
ldnd,is 130 eV 2c MnKa {589() eV).The detector re,ohmon a funrnon of X-ray ener~y .rnd it i~ about
65 eV at 282 eV.'Jltere ,n: sevtraJ advanced detecton ~uch a, rht' \llt<.011 dnfc <let:tt.•rcor, "hid1 \\mks at
250 K, having the s:une rt~oluriou but an mcrcased s;tmplmg rate. A m1croc.1lorinwtt•r U)\ \\llrks b)
measuring the temper.Hurt: differtnce camcd by the absorpnon of X ra>, WHh thi', un 111cre.hc.>d ent"rg)
resolu □on is possible, wluch 1> about 4 5 eV at MnKa.
Wavdengp, dispersive spec1rcm1t:te~ work 011 the pr•11( 1plt' of Br.1gg ditfo1c1io11, ;i,l ::: 2.d,1118, "hC're
A. 1~ the w,velength of rht X- r.1y .111 d 0 I\ the J11glt' of 111t 1dt•11t l' l ht'1l'lun-, .11 \ ',1mHI\ ,111gll♦, lit 111c1dcnce,
tt 1~ pos~ible to scatter rays of diffortllt wJvde11htth~ to thr dett't ma A, ,1110 l .llllllll ht' l;.u~c•r tlun one", the
long~t wavelength diffracted h 2d, thcrcfon· the t I y~t.il put, ,1 ltm11 011 tlll' 1.111gl' llf l'lt·mt'nh l li,-en•d
There are also other limfrations of the han.Jwa,e. Wlieu O 1, 'JlJ'>, dtt' dt•tt•no1 h.1' ll 1 ,u ,It the X ray
source; when sm 0 15 close to zero, the movcuitut of tltl' t 1) ,c.tl 11t-.11 tht• \Pl'l llllt'II 1, ,1 l*'bll·m. In order
to diffiact the low energy X-rays emitted by ilc, B, C .111d N, it h lll'l'l'\\,11 y ru 11\l' t l r~t.11~ of l.1rgl' d, alues.
34
0
.,, ~ poss h c b\ L:mgmmr-Biodg ett films (,ee che chapter on )elf-assembled 'Monoldyers) L
h
~ f f ' s p ~ an be CW1ed br ,-an,ng che ch.am len!-rth of the monofaver. ;\ltosr of th \\· oSc
-i-~ • d , e measure
ot - -":n ...-A-Ue~ X-r.n-s are one b,· • Ln-ered
• . svnthetic
• . microscrucrures .This co n ca·ms evaporated alteIllents
tn-cn ct ht:-l'-.. .md light demems. The duckne:5.s ot the byers can be e.isiJy controlled. ~te
Th: 5P4ml resolution I ;·SE\I is ~ecred by the beam size as menc:ioned before. Bue the beam causes
th ch . .
m:~-oon w·1 e ~ ~ ...n..l the region of interaction exceeds che physical dimension of th b
mcre-..ses J..uer.tllv as ,veil as depth-wISe, and .L:. c ·
ui.u m eracnon vo1ume
e eam
d epends
:rad:. The
J.._ d _e~ent
--=-•cf>4tul __ • on
w:r nu:cu.i • an 1
ii.DU • on .he beam enerP"\,•
_ •::,r i \r larg · th b ·
e energ1es. e earn mceracr.s to a greater extent, while
ch-: dqd1 ~f 1m·eacno~ d ~ ~,~ch a decrease in elecrroo energy till about 100 eV and at ve low
energies o.: die o.-der o.r a ~e'\'-" e \~ It mcreases again. The range of interaction is decided by the e17ctron
~ ~ o n 1en;-...h. The uruversa.I electron acreouac:ion lengrh of materials IS given m Fig. 2.7. As seen m
me tr_gu.:-e. die a~e-nw.non leng-..h increases "ich electron energy and has a minimum at a few hundred v
This nu.k~ ir possible co design highl·.- surface-sensitive SE.\11.s by reducmg the beam energy. T~
~--dopmen~ is discussed in low energ", SE\.1.