Scanning Electron Microscope Lab Report

Report presentation on the Scanning Electron Microscope, for the Methods and Instruments of Analysis course, in the Nanotechnology MSc program at the Information and Telecommunication (ICT) school KTH.

1. IH2652 Methods and Instruments of Analysis
Scanning Electron Microscope
Lab Report
December 2, 2014
Group members: Sumit Mohanty
Mohamed Atwa
Ahmed AlAskalany
1

2. Contents
1 Introduction 3
2 Results and discussions 3
2.1 Acceleration Voltage and Charging Eect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Current Eects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.3 Tilt Eects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.4 Working Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Conclusion 6
2

3. 1 Introduction
Scanning Electrons Microscope (SEM) is a powerful characterization tool for materials of various sizes and
shapes, and is extremely important in the study of nano-scale structures. The electron beam from an electron
gun is directed towards the specimen and steered using an arrangement of apertures and magnetic lenses.
The steered electron beam is focused on specic positions on the surface of the specimen, where dierent
beam-sample interactions can take place, the electrons resulting from such interactions are collected by spe-
cic detectors and converted to an image on a screen. There are two important types of electrons resulting
from beam-sample interaction. The rst is the Back-scattered Electrons (BSE), which result from elastically
scattered of incident beam electrons by atoms in the sample, BSEs are used to generate images with contrast
varying with the atomic number of dierent sample constituent elements. The second are Secondary Elec-
trons (SE), resulting form inelastic scattering process, these are useful in obtaining topographical images
of the sample's surface, since their reected intensity is proportional to the volume of interaction that is
aected by the angle between the beam and the point where it hits the sample.
Various parameters aect SEM images, accelerating voltage, beam current, beam diameter, working distance
are all parameters that should be taken care of in order to obtain a high quality SEM image. SEM images
always seem to be illuminated from a certain direction, which is actually the detector direction.
SEM samples should either be conductive or coated by a conductive layer, which can be done by dierent
techniques. This is to eliminate any accumulation of electrons on the sample's surface (surface charging)
degrading image quality. In case of biological samples, it is required that water is replaced by appropriate
gas, to avoid water bursting under the inuence of electron bombardment.
In the lab, dierent SEM images were obtained, examining the dierent eects of chaning accelerating
voltage, probe current, tilt angle, and working distance on the SEM image. In this report, the dierent SEM
images are presented and accompanied by comments on the dierences based on the theory of operation of
SEM.
2 Results and discussions
In this section of the report the dierent experiments conducted in the lab will be discussed, and will be
commented on based on the theory of operation of SEM and the possible interpretations of the eect of
change in its dierent operating parameters on the obtained scans of the specimen.
2.1 Acceleration Voltage and Charging Eect
The eect of varying acceleration voltage was studied at a constant magnication of x5k. It could be clearly
seen that the image obtained at highest value of acceleration voltage (Fig. 1d)of 20kV is the brightest. Its
reasonable to argue to that this primarily owes to the highly energetic electrons bombarding the sample
thereby and producing a more illuminated image. Hence this overtly illuminated image might lead to the
charging eect of the specimen which may damage our sample. These electrons have a lower wavelength
thereby a higher energy which leads to higher electron beam brightness and ultimately a higher resolution.
Similarly, it can be seen that as we go down the values of acceleration voltage the image, we lose this bright-
ness and henceforth at the value of acceleration voltage as low as 1kV, it is dicult to obtain a clear image
even by manually enhancing the brightness on the screen. The image obtained as a result was distorted as
shown in Fig 1(a). Hence, a higher acceleration voltage reveals more surface details on a specimen due to
higher resolution.
On the contrary, a brighter image doesn't necessarily imply more clarity because although a more brighter
image is obtained upon increasing the acceleration voltage, the contrast between the features isn't satis-
factory. Increasing the acceleration leads to an increase in the interaction between the electron probe and
specimen in the lateral direction which leads to a decrease in resolution. There lies a trade o between using
the highly energetic electrons for illumination and stark contrast between the intricate structures apparent
from the gures. Looking at Fig 1b and 1c, it can be observed that the miniatured hole like features on
3

4. (a) SE image 1kV x5k(b) SE image 5kV x5k(c) SE image 10kV x5k(d) SE image 20kV x5k
Figure 1: Multiple SE images corresponding to dierent acceleration voltages
the electrode ngers have much more clarity compared to the two extreme values and therefor these two
represent a more optimal case of tuning between the brightness and contrast and ultimately the resolution.
Therefore a right balance between brightness and contrast could only be obtained at a value roughly between
between the two extrema for the perfect resolution.
The eect of varying the accelerating voltage on SE images can be summarized as follows. Increasing the
accelerating voltage causes a considerable increase in image resolution, on the expense of the clarity of surface
structures, also it leads to more charge-up, and increased probability of sample damage. While decreasing
accelerating voltage lowers the image resolution, however it allows us to see surface structures more clearly,
and reduces other eects associated with higher voltage, like charge-up and damage to the sample.
2.2 Current Eects
(a) 20mA(b) 40mA(c) 60mA
Figure 2: SE images at 20kV and dierent beam currents, to study how current intensity aects image.
In conclusion, increasing probe current generates a smoother image with deteriorated resolution, and may
cause damage to the sample. While a higher resolution image with less probability of damage can be obtained
using lower probe currents, but the image will be grainy due to the decrease in the signal-to-noise ratio.
4

5. 2.3 Tilt Eects
(a) 0 degree tilt(b) 45 degree tilt
Figure 3: The eect of changing the tilt of the sample on the SEM image
In order to look at the tilting eect of sample, the working distance was increased to roughly 27mm to
accommodate the angular displacement of 45
o
, this increased the working distance from the perspective
view and hence the image obtained are shown in Fig 3. Fig 3b makes it apparent that the sample is
disproportionately illuminated. This is because of the non uniformity of the electron beam irradiation on
the specimen and the relatively more lit up area corresponds to the direction in which the detector perceives
it. Therefore, increasing the tilt apart from obvious perspective scaling of dimensions, also results into
disproportional irradiation of the specimen resulting into variable resolution along the sample. This is
analogous of having dierent working distances at dierent points of the sample.
2.4 Working Distance
Figure 4: Working distance changed from 7mm to 33mm
A variation of working distance (WD) was carried out from 7mm to roughly 33mm as shown in Fig 4.a-4.c.
It could be observed that moving to closer WDs gives a higher resolution of images as apparent from the
image 4c. Though there again lies a trade o between the resolution and depth of eld. Looking at image 4a,
it could be seen that the depth of eld is improved but at expense of poor resolution. Furthermore, it greater
WDs implies imparts a certain fringing eect to the image which even worsens the resolution. This fringing
eect owes partly to the improper spherical aberration and also to the inconsistency in the rate of electron
scanning. There is a possibility that at higher WDs, the rate at which electron beam scans the sample has a
certain lag (owing to the distance) compared to the data acquisition rate. The variation of this lag could be
responsible for the fringing eect (Fig 4a) which looks like an interference pattern on the electrodes in the
image. On a conclusive note, the optimum balance of the working distance could be obtained, depending
upon the requirements of depth of eld and resolution, somewhere in the intermediate values of WDs chosen.
5

6. 3 Conclusion
During the lab, dierent experiments were conducted. In the rst experiment, dierent SEM scans were
obtained for the specimen, at dierent acceleration voltages (1kV, 2kV, and 5kV) to study the charging
eect of the specimen on the SEM image, and how this can aect the image's resolution. In the experiment
that followed, the eect of varying SEM beam current on the image was examined through setting the
beam to a low (20mA) and high (40mA) values. The eect of dierent beam currents was discussed, in
terms of image resolution, grainy image, signal to noise ration, and the possibility of sample damage due
to charging eect. Afterwards, the sample was scanned before and after a tilt of 45 degrees, with same
operating conditions. This was discussed in terms of the disproportional reection of the irradiated parts
of the specimen, depending on their perspective with respect to the detector. Finally, the working distance
of the sample was increased, and the eect of this on the image's resolution and the depth of eld was
explained.
The most important part of this lab, besides learning about the eect of the change in dierent operating
parameters, was getting an important hands-on experience and having to learn how to operate an SEM and
to tune its dierent variables in order to obtain a clear image that is essential for the study of dierent
materials and components.
6