【求助】求助---哪位高手大侠可以指点下：环境扫描电镜的工作原理是什么

环境扫描电镜的工作原理和扫描电镜的工作原理是相通的，只是探头工作的真空环境变化的

谢谢
本人对它在低真空下怎样由电子束激发出信号，以及探头怎样能在低真空下收集信号不太明白
可否具体说明一下，十分感谢！！！

针对扫描电镜对潮湿及易挥发样品的分析局限，ESEM的出现刚好解决了这方面的难题。ESEM最大的优点是在于它可以改变显微镜样品室的压力、温度及气体成分。它不仅保留了SEM的全部优点，而且消除了样品室必须是高真空的限制。潮湿、易挥发和不导电样品在自然状态都可以进行观察，无需进行任何处理。在气体压力高达5000Pa，温度高达1500℃    ，含有任何气体种类的多气体环境中，ESEM都可以提供高分辨率的二次电子图片，从而使SEM的使用性能和应用范围有了很大的改善。
ESEM同样可以与X射线能谱仪相配接，进行元素分析，材料元素的面分布图和线扫描曲线，即使对超轻元素。分析精度也不受影响。

我敢和楼主打赌，对于您的问题没有几个人能回答清楚，大多数答案都会是模棱两可，似是而非的，或像1楼那样装傻充愣，或者干脆是像3楼一样销售之谈。

是否精确性的原理很少有人知道？！

There are now two distinguishable varieties of SEM with the ability to operate with the sample in gas. The Environmental SEM is a trade-mark of Philips/FEI/Electroscan and refers to an instrument which can operate with a pressure in the specimen chamber of up to 2,700 Pa (20 torr) and which offers a proprietary secondary electron detector for use in the gas. Instruments from other manufacturers are usually referred to as variable ¬pressure SEMs (VPSEMs), nature SEMs, or some similar term. Typically, although not exclusively, such machines are limited to a maximum pressure in the specimen chamber of about 266 Pa (2 torr) and may not offer a secondary electron detector for use when there is gas in the chamber, relying only on BSE detection for imaging. However, the similarities among all of the instruments are more important than any differences and so their design and operation can be discussed without reference to any particular version.
Typically these instruments use a relatively modest electron-optical system and a basic thermionic gun because the performance of the machine is ultimately limited by the electron-gas interaction, discussed below, rather than by the lenses. Instruments employing field emission guns and more advanced lenses are now being built to offer higher levels of resolution. VPSEMs and ESEMs are also usually designed to operate optimally in the energy range 10-30 keV rather than at the lower energy of 1-5 keV favored by conventional high-vacuum machines. In all other respects, however, VPSEMs and ESEMs and conventional SEMs are identical.
These instruments offer significant advantages when compared to con¬ventional SEMs. The relaxed vacuum environment means that many sam¬ples that would otherwise be unsuitable for observation, for example, ma¬terials such as bone that are porous and hard to pump down, specimens that are damp or wet, or biological specimens that cannot be maintained in their original state if they are allowed to dehydrate, can be imaged safely and conveniently. The gaseous environment allows poorly conducting and insulating samples to be imaged in a stable manner at any desired beam energy without the need to coat them with a conducting metal layer. The gas in the specimen chamber also creates a microenvironment which can be used for a wide variety of in situ experiments (e.g., in corrosion and oxidation). Moreover, the gas provides new modes of imaging, which can generate new types of information (Griffin, 2000). Finally, with minor mod¬ifications, these instruments can also operate as conventional high-vacuum .SEMs and so offer great flexibility to the user who is confronted with a wide variety of problems and specimens.

When the pressure is in the I0-4-Pa (l0-6-torr) range, typical of the vacuum level found in a conventional SEM, then the electron mean free path is of the order of 10 km and so the probability of any given electron being scattered on its way along the roughly I-m path from the gun to the specimen is negligible. However, when the pressure is raised to 100 Pa, the mean free path drops to about 1 cm, which is of the order of the working distance of a normal SEM. For still higher pressures the mean free path becomes 1 mm or less. The effect of this change in the mean free path can be understood by considering a simple model. The number of collisions m suffered by an electron traveling a distance s when the gas mean free path is λ    is m = s/λ. Because the distance traveled by each electron in the VPSEM or ESEM is the gas path length (GPL), the average number of collisions per electron m = GPL/λ. It can also be shown that the chance of an electron reaching the specimen without being scattered at all is exp(-m). Depending on the value of m, different imaging situations are encountered.
If the chamber is maintained at high vacuum (≤10-3 Pa), then for a normal working distance or gas path length of 15 mm, m will be 0.0001 or less. In this case essentially all of the beam current leaving the lens will reach the specimen, and the beam profile will remain undisturbed because scattering events will be rare. This is the situation encountered in a normal SEM. If the chamber pressure is high, however, for example, 400 Pa (~3 torr), then the mean free path shrinks 5 mm and so for a gas path length of 15 mm, m now has a value of 3. In this case only about exp( - 3) = 0.05 = 5% of the beam is unscattered, and the focused beam is almost completely destroyed, so scanning electron microscopy would be impossible. If, however, the path length is reduced to a few millimeters or the gas pressure is lowered to 100 Pa or so, then the value of m becomes about unity. In this case a significant portion of the beam [exp(-l) = 0.37 = 37%] remains unscattered, and a focused probe is still formed, although it is surrounded by a "skirt" of scattered electrons. The effect is analogous to that of viewing a car headlight through fog, where the main beam of light would still be visible, but would be surrounded by a halo or skirt of scattered light. Figure 5.19 shows a visualization of this kind of beam (Wight and Zeissler, 2000) obtained in an ESEM. The very sharp and intense central focused beam can be seen to be surrounded by a broad and less intense skirt region extending for a significant distance away.

Imaging in the ESEM and the VPSEM
The incident beam of electrons hitting the surface of the specimen in the ESEM or the VPSEM produces the same result as in the high-vacuum SEM, so, in principle, the usual backscattered and secondary electron signals are available. However, although there is no problem in collecting the backscat¬tered electron signal, the secondary electron signal is not available because, as noted above, the SE only travel a very short distance from the sur¬face of the sample before they interact with and ionize a gas molecule and so it is difficult to collect them. Moreover, the conventional E- T detector used for SE collection relies on a high (+ 10 kV) bias voltage applied to the scintillator to accelerate the SE to sufficient kinetic energy to stimulate light emission from the scintillator. In the high-gas pressure environment of the specimen chamber, where a potential of more than just a few hundred volts would cause the gas to ionize and the detector to arc over to ground, the E- T is thus not a viable option. It is thus impossible to collect a SE sig¬nal in the conventional manner. Because images prepared with a positively biased E- T detector (consisting of SE and BSE contributions) account for more than 90% of all SEM images, this is clearly a problem.
VPSEMs switch off the E- T detector when there is gas in the chamber and rely on the passive backscattered electron detector. At first sight this may appear to be a poor exchange, but in fact for many purposes a BSE detector is a quite acceptable substitute route for imaging. First, VPSEMs are typically used at beam energies of 10 keV and higher (to minimize beam broadening) and for almost all materials the yield of BSE is higher in this energy range than the SE yield. Because modem BSE detectors are very efficient, the quality of the BSE image is in every way comparable with the corresponding SE image. Second, most VPSEM operation occurs at magnifications below 20,000 X. In this operational range the pixel size of the image is of the order of a fraction of 1 um or more, and consequently the information in the SE image comes predominantly from the SE2 electrons (i.e., those secondaries generated by the backscattered electrons) because the generation volume of the SE2 signal is comparable with the pixel size. The SE2 and the BSE signals are therefore almost identical in informa¬tion content and resolution because one is essentially a copy of the other (Fig. 5.27). This does not necessarily imply that the images will look identi¬cal because the BSE and SE detectors are usually placed in rather different locations relative to the specimen (remember that the apparent source of illumination always comes from the detector), but it shows that there is no fundamental loss of imaging capability in this condition. Finally, be¬cause BSE detectors are much less sensitive to charging effects than are SE detectors, the use of a BSE detector in the VPSEM simplifies the challenge of overcoming charging. The strategy of providing only a BSE detector in a VPSEM is therefore not a major limitation to what can be achieved in the instrument.

Nevertheless there are instances where a true secondary image is de¬sirable, for example, to image the actual surface of a specimen, and so in the ESEM and in some VPSEM instruments, special devices known as environmental SE detectors (ESED, Danilatos, 1990) or gaseous SE detec¬tors (GSED; Farley and Shah 1991; Mohan et al., 1998) are now provided to fill this need. Although the exact details vary from one system to an¬other, the fundamental principle behind all of these devices is identical (Fig. 5.28). As seen above, it is the secondary electrons that are respon¬sible for the majority of the gas ionization, so the amount of ionization varies with the SE yield: In the presence of an electric field, produced by positively biasing an electrode above the sample, the SE are acceler¬ated and eventually gain enough energy (10-30 eV) to ionize another gas molecule. The two electrons that result from this process are again ac¬celerated and each in turn may generate two more electrons and so on, producing an avalanche effect known as gas amplification. The magnitude of the amplification depends on the ease of ionization for the chosen gas and on the field available. With potentials of the order of a few hundred volts (low enough to avoid electrical breakdown of the gas) and electrode¬-to-sample distances of a few millimeters, the electron current collected on the biased electrode is hundreds or thousands of times larger than the original signal, which is a fraction (5-20 %) of the beam current. The positive ions produced in this cascade are themselves accelerated away from the positive electrode toward the specimen, forming an ion current equal in magnitude to the current of electrons moving in the opposite di¬rection. For the same kinetic energy, ions move much more slowly than electrons because of the great difference in mass (a factor of 1837 per proton equivalent, or about 33,000 for an H2O+ ion). The slowly moving ions determine the time-dependent response characteristics of this class ofdetector. .
The electrode and the sample effectively form the two plates of a ca¬pacitor with the gas as a dielectric, and the drift motion of the ions between these plates induces a current in the external circuit, which can then be amplified and used to produce the required image. It is not necessary for the ions (or even any residual electrons) to be physically collected by the detector because the motion of any charged particles between the plates will result in a signal current (Ramo, 1939). Because the initial source of the ionization was the secondary electrons generated by the incident beam, the signal from this system is in every respect a secondary elec¬tron image, although it was carried by ions. A comparison of high-vacuum SE images (Fig. 527a) and the corresponding images generated by ESED or GSED systems in gas (Fig. 5.29) confirms that the images are iden¬tical except for changes that can be attributed to the different positions of these detectors relative to the specimen. There is, However, one prac¬tical difference between ESED/GSED images' and those recorded using an E- T detector. Because ions travel much more slowly than electrons of similar energy, unless slower scan rates are used the image appears blurred because information from one pixel dwell time period is still being collected when the beam has already stepped onto the next pixel (Newbury, 1996).
The ESED and GSED detectors give ESEM and VSPEM instruments all of the imaging power and flexibility of the conventional E- T detec¬tor, and they also provide other benefits. Because surface charging can always be eliminated by the gas, the images that are obtained do not show the distortions, instabilities, and extremes of contrast due to charging that are encountered on conventional images. An excellent example of this (Fig. 5.30) are images from deep holes in a surface (Newbury, 1996). In a conventional SE or BSE image the bottom of the hole in an insulator is dark because charging on the walls of the hole traps electrons and prevents them from leaving. In the presence of gas, and when using an ESED or a GSED system, however, not only does the absence of charging allow electrons to leave freely, but the added gas path length experienced by the ions that are generated increases the amplification with the result that the bottom of the hole is actually brighter than the surface itself.
An additional benefit of the ESED/GSED detectors is that, unlike the conventional E- T detector, they are not sensitive to light. They can therefore be used to image samples that are at a high temperature (i.e., above about 750°C, where the specimen begins to visibly glow) without the annoyance of a strong and noisy background signal. This ability is exploited in the ESEM by the provision of stages capable of taking specimens to 1500°C or even higher.