ROUGH DRAFT authorea.com/92669

# Environmental Electron Scanning Microscope

Abstract

This is the term paper for the course AEP 6610, reviewing the techniques and development of environmental scanning electron microscope (ESEM). The ESEM provides the ability to image samples down to nanometer scale without the necessity of high vacuum in the chamber and sample processing procedures. Therefore, samples can be imaged in its original hydrated state, preserving its dynamics, interior structures and morphology. Secondary electrons are collected to reveal the topology of the sample while backscattered electrons are collected to distinguish the element composition. In this paper, the resolution and limitations for ESEM are presented. The applications of ESEM on both organic and inorganic materials are discussed. Finally, future prospects an comparison with competing imaging technologies conclude the paper.

# Introduction

The ESEM(Stokes 2008) was developed from conventional scanning electron microscopy (SEM)(Reimer 1978), preserving the principles of imaging and the resolution. The advancement and modification of SEM simplified the preparation of samples and allowed more complex imaging environment. The conventional SEM is a surface analytical technique; it has to be operated in high vacuum ($$\approx 10^{-5} \sim 10^{-7} \space Torr$$) chambers to prevent surface contamination. Therefore the samples must be clean, dehydrated, fixed and also conductive to avoid the charging effects. Toward these purposes, the samples are mostly pre-processed such as cooled, dehydrated and distorted, and usually coated with conducting materials, compromising the topographical and morphological resolution. The lifetime of samples for SEM is also short. The ideal samples for SEM are metal surfaces. However, in ESEM, the samples can be both organic or inorganic and conductors or insulators which are imaged under a range of pressures, temperatures and gases. The problematic issues in SEM such as contamination, short sample lifetime and charging effects can be handled in ESEM.

ESEM consists of similar components as SEM. The ESEM components are shown in Fig.1(Donald 2003). An electron chamber that sits on the top of the sample chamber contains a heated filament, accelerating anode, condenser lenses and objective lenses. Between the electron and sample chamber, there are two stages of environmental chambers. The sample chamber is usually contains different gas molecules and is kept at certain higher pressure from $$0.1 \sim 30 \space Torr$$ compared with the high vacuum required in SEM. Each chambers are kept at different pressures by vacuum pumps and separated by multiple pressure limiting apertures (PLA’s). The functions of PLA’s will be discussed later in this section.

A schematic components representation for ESEM. The chambers are kept at different pressures by differential pumping.

## Electron Chamber

The electron chamber is divided into two parts; one is the electron gun and the other is the column lining chamber consisting of many magnetic lenses. Electron beams are generated in the electron chamber by a filament heated to high temperature (thermionic emission) or kept at strong electric field (field emission). The electrons emerging from the filament are then directed through a small spot before being accelerated in the electron gun toward the column lining chamber. The accelerating voltage is typically in the range of hundreds to tens of thousands of volts. The electron gun must be operated in very high vacuum ($$10^{-7} \space Torr$$) conditions to minimize scattering. The electron beams enter the column lining chamber as divergent and broadened beams; therefore, condenser lenses are required to converge and focus the beam on a small crossover and toward the two environmental chambers.

## Environmental Chambers and PLAs

One of the key development in ESEM is the chamber design that keeps high vacuum at the electron gun and relatively low vacuum in the sample chamber at the same time. The pressure gradient from $$10^{-7}$$ to $$30 \space Torr$$ is achieved by multiple stage of chambers with tiny opening to other stages. Such stages are the two environmental chambers (EC1 ad EC2) which separate the column lining chamber from the sample chamber, functioning as buffer zones to minimized the pressure fluctuation. The chambers are connected by two pressure limiting apertures (PLA’s). The PLA has a tiny opening for electron beams, through which the gaseous molecules will leak out from high-pressure side to the other. Even though the opening is designed as small as possible, serious leakage takes place without the differential vacuum pumping. As the gaseous molecules leak through one PLA from pressure $$P_2$$ into the gap between two PLAs maintained at pressure $$P_1$$, they will be removed from the gap region by a vacuum pump. Therefore the pressure $$P_1$$ remains intact. However, the molecules in the gap will also leak into the chamber with pressure $$P_0$$ through the second PLA. Another more powerful pump is required for $$P_0$$. The flux $$\Phi$$ of the gaseous molecules is the number of molecules $$dN$$ flowing through an area $$dA$$ within certain period of time $$dt$$. The flux in the opening of a PLA is proportional to the pressure difference between the two sides of a PLA. $\Phi = \frac{dN}{dA dt} \propto \frac{P_H-P_L}{P_L} = \frac{\Delta P}{P_L}$ where $$P_L$$ and $$P_H$$ are the lower and higher pressures respectively. If $$P_L \ll P_H$$, $$P_H-P_L \approx P_H$$. The flux is then proportional to the ratio of $$P_H$$ to $$P_L$$. $\Phi = \frac{P_H}{P_L}$ Therefore, the flux from $$P_2$$ to $$P_1$$ and from $$P_1$$ to $$P_0$$ can be written as the following expressions with a proportional constant $$\alpha$$. $\Phi(2\to1) =\alpha \frac{P_2}{P_1}$ $\Phi(1\to0) =\alpha \frac{P_1}{P_0}$ If, say, the pressure of the sample chamber $$P_2=0.1 \space Torr$$, the the pressure of the column $$P_0 = 10^{-5} \space Torr$$, take the pressure in the gap of PLAs to be the geometric average of $$P_2$$ and $$P_0$$, i.e. $P_1 = \sqrt{P_2 P_0} = 10^{-3} \space Torr$ The leakage flux into the higher vacuum chamber is $\Phi(1\to0) =\alpha \frac{P_1}{P_0} = 100 \space \alpha$ Compared with the situation where only one PLA is applied to separate the sample chamber and the column, by applying the same formula, the leakage flux is $\Phi(2\to0) =\alpha \frac{P_2}{P_0} = 10^4 \space \alpha$ The introduction of the second PLA and the buffer zone between the two PLAs reduces the leakage by 2 order of magnitude. Still, this leakage effect is significant if there is only one buffer zone. Experimentally, more PLAs are used to establish more buffer zones and minimize the effects of gas leakage. Typically, the two ECs are maintained at about $$10^{-4}$$ and $$10^{-2} \space Torr$$ respectively.

## Sample Chamber

The sample chamber provides a highly flexible environment where the samples can be preserved and imaged under situations as original as possible. The pressure in the sample chamber ranges from $$0.1 \sim 30 \space Torr$$. The pressure is controlled by a gas inlet to complement the gas leakage. Water vapor is preferred as the gas source since the samples are usually hydrated. The temperature of the sample chamber can also vary wildly from cryogenic samples at $$120K$$(Meredith 1996) to in-situ uranium-cerium mixed oxalate at $$1500K$$(Podor 2012).