Note: O-ring stack system for electron gun alignment In-Yong Park,1 Boklae Cho,1 Cheolsu Han,1 Seungmin Shin,1 Dongjun Lee,1 and Sang Jung Ahn1,2,a) 1 2

Korea Research Institute of Standards and Science (KRISS), Daejeon 305-340, South Korea Major of Nano Science, University of Science and Technology, Daejeon 305-340, South Korea

(Received 4 November 2014; accepted 26 December 2014; published online 15 January 2015) We present a reliable method for aligning an electron gun which consists of an electron source and lenses by controlling a stack of rubber O-rings in a vacuum condition. The beam direction angle is precisely tilted along two axes by adjusting the height difference of a stack of O-rings. In addition, the source position is shifted in each of three orthogonal directions. We show that the tilting angle and linear shift along the x and y axes as obtained from ten stacked O-rings are ±2.55◦ and ±2 mm, respectively. This study can easily be adapted to charged particle gun alignment and adjustments of the flange position in a vacuum, ensuring that its results can be useful with regard to electrical insulation between flanges with slight modifications. C 2015 AIP Publishing LLC. []

Charged particle beam instruments (CPBIs) have been used extensively in both industry and science in the areas of semiconductors, micro/nano-scale machining, material analysis, and in biotechnology, etc. There is currently much effort being expended with regard to the development of a better imaging and fabrication resolution, as high-performance CPBIs are continuously required. There are several factors related to improved performance. Among them, the exact alignment between each optical component is very important. As a general rule, CPBIs consist of a source or a gun, a column, a sample holder, and a detector part. These optical elements have to be aligned precisely with one another in order to gain a high resolution while also reducing aberration effects.1,2 For a perfect alignment, the source position and the beam direction between the gun and the column are crucial. A precise alignment is typically accomplished with the following two methods. The first involves a mechanical adjustment that physically shifts each component using screws.3 The second relies on a double-deflector, which tilts and shifts the beam with an electromagnetic field.4,5 Most mechanical alignment operations use sliding O-ring seals or thin metal bellows, which allow motion while preserving the internal vacuum. However, only changes of the shift and small tilting angle are possible with an O-ring, and it is difficult to modify the metal bellows depending on the required alignment range. For the doubledeflector, an electronic part controlling the applied voltage or current is necessary. Moreover, assembling double-deflector part inside a vacuum chamber is fairly difficult. A vacuum condition along the charged particle beam path is essential to prevent collisions and scattering of the charged particles with air. Therefore, the alignment system should allow simple and highly accurate motion freedom while maintaining a stable vacuum condition inside the system.

a)Author to whom correspondence should be addressed. Electronic mail:

[email protected] 0034-6748/2015/86(1)/016110/3/$30.00

In this investigation, we realize a proper electron gun alignment with a stack of Viton O-rings which can be tilted and shifted mechanically without any type of electronic control. The characteristics of this system are measured by means of beam imaging on a phosphor screen. Figure 1 shows the general correction steps required for a misaligned beam between a gun and a column, somewhat exaggeratingly. Regardless of how well the gun and column are designed and assembled, the beam direction is generally not aligned along the optical axis, as shown in Fig. 1(a). As discussed earlier, misalignment causes distortion and aberrations in the beam shape, finally inducing poor imaging and milling resolutions. For that reason, shift and tilt motions are applied to the alignment procedures. These functions should be implemented during the development of CPBIs and the replacement of source parts such as the tip and filament.

FIG. 1. Brief description of general correction procedures for misaligned beam. (a) Misaligned beam between gun and column. (b) Gun part is shifted to optical axis linearly. (c) Finally, gun part is tilted and the beam direction is parallel with optical axis.

86, 016110-1

© 2015 AIP Publishing LLC


Park et al.

Rev. Sci. Instrum. 86, 016110 (2015)

the beam and the phosphor screen (L = 670 mm). We adjust the tilting angle of the gun flange through a height difference, ∆V , at the completely opposite position V1 and V2. From this simple relationship, the tilting angles along the x and y axes are estimated as follows: ∆Vx, y . (1) D Here, θ x, y denotes the tilting angle along the x and y axes with respect to the original beam direction. In addition, Mx, y is calculated comprehensively as the moving distance of the beam imaging on the phosphor screen along the x and y axes by tilting. This is expressed as follows: θ x, y = tan−1

Mx, y = L × tanθ x, y .

FIG. 2. Schematic diagram of the electron gun and adjustable O-rings system. Filament current for generation of electron beam and Wehnelt voltage (Vext) for beam extraction are floated on an acceleration voltage (Vacc). The anode and phosphor screen are connected to the earth as a ground directly. The beam is rotated at the pivot point and the M means the moving distance of beam pattern at the phosphor screen.

Figure 2 shows a schematic diagram of the apparatus to be tested in this work by an electron beam. We utilize a thermionic electron emission source which produces a flow of electrons from a heated tungsten filament for which the outer orbital electrons gain enough energy to overcome the work function energy. We maintained a vacuum pressure of approximately 10−5 Pa with a combination of a rotary pump and a turbomolecular pump which ran in tandem. In order to accelerate the generated electron beam, the electron gun system is floated at a negative 20 kV (Vacc) and the filament is heated by adjusting the current flow. Given that the electrons are emitted in all directions from the heated tungsten filament, we used an accelerating lens system, which consists of two anodes to control the beam direction and shape. The first anode, which is generally referred to as a Wehnelt cap, is positively biased (Vext) against the floating ground while the second anode is connected to the earth as a ground. Therefore, the electron beam passing through the Wehnelt aperture is accelerated into the second anode due to the potential difference between the two anodes. The phosphor screen is also connected to the earth ground so that the electron beam passing through the second anode hits the phosphor screen at an energy level of nearly 20 keV. This makes it possible to observe the imaging of the beam with only the phosphor screen and without a micro-channel plate which amplifies the original signal. For an analysis of the tilting mechanism along the axes, there are four key parameters of the system: the diameter of the flange (D = 152 mm), the two flange-to-flange distances in the O-ring stack (V1 and V2), and the distance between the pivot point of


Also, the linear shift of the gun is of particular importance with regard to the beam alignment. In relation to this, we move the gun flange along the horizontal axis by tightening and loosening horizontal clamping screws. H1 and H2 mean the distance between a screw support and a flange at the completely opposite positions. For vertical shifting, we adjust the vertical clamping screws while leaving unchanged the relative height at four places where the clamping screws are in contact with the flange. We assembled four identical structures, all of which can shift the gun flange vertically and horizontally with clamping screws and separate it azimuthally with an equal space on the flange, which is under the O-rings, as shown in Fig. 3. The system should maintain a steady vacuum pressure inside the chamber and should be structurally sound during the beam alignment process. Thus, we add center rings between the Orings when the latter are stacked. By doing so, the structure of the O-ring stack remains stable during the tilting and shifting of the flange. We use clamping screws which have a ball bearing at the end of the screw so as to make the flange move smoothly at the contact point. Considering that the tilting angle is related to how much the O-ring is compressed relatively, we can readily regulate the full range of the tilting angle with the total number of O-rings. As shown in Fig. 3, we also designed and assembled a structure to allow variations in the height according to the number of O-rings. Therefore, we can modify the O-ring stack easily and quickly according

FIG. 3. Picture of the adjustable O-ring stack system.


Park et al.

Rev. Sci. Instrum. 86, 016110 (2015)

FIG. 4. Electron beam pattern observed with the aid of phosphor screen and digital camera by tilting a flange of electron gun. (a) Accumulated consecutive snapshot images with tilting the gun along the x axis. (b) Accumulated consecutive snapshot images with tilting the gun along the y axis. Dotted lines in (a) and (b) indicate beam images captured before tilting.

FIG. 5. Overlapped beam images captured by shifting a flange of along the x and y axes using a phosphor screen and a digital camera. (a) Electron gun shifting along the x axis. (b) Electron gun shifting along the y axis. Dotted lines in (a) and (b) indicate beam images captured before and after shifting linearly.

to the maximum tilting angle. To ensure that the vacuum pressure remains stable, we assume that the each O-ring can be compressed by approximately 1 mm in combination with the center ring. Therefore, using the condition of ten O-rings in Eq. (1) means that the maximum value of θ is approximately ±3.76◦ theoretically. A sequence of snapshots of the beam image is shown in Figs. 4 and 5 to explain how the position of the beam changes on the phosphor screen with the tilting and shifting of the flange. In order to create an accelerated beam pattern on the phosphor screen with high visibility, we set the Wehnelt bias voltage to +0.544 kV. Hence, the electron beam current was found to be approximately 32 µA. In order to observe this well, we overlap each beam image which is captured after ensuring a height difference along the x and y axes in steps of 1 mm, as shown in Figs. 4(a) and 4(b). As shown in Figs. 4(a) and 4(b), the beam pattern moves well in parallel motion along each axis, indicating that we can tilt the beam direction precisely on two axes, as intended. However, the screws and the flange are only in contact without any guides or joints. Accordingly, the x and y axes tilting correction could affect a little each other, as shown in Fig. 4(b), which shows some very slight curvature. In Figs. 4(a) and 4(b), the maximum compressed value of ∆H x, y is 7 mm in the experiment, and this height difference is theoretically converted to ±2.64◦ after calculating it with Eq. (1). If we calculate the maximum tilting angle with L and the beam moving length (Mx, y = 33 mm) from the experimental result as measured on the phosphor screen, the tilting angle is then estimated to be about ±2.55◦. This is nearly equivalent to the theoretical calculation, indicating that we can control the gun tilting angle precisely, as intended. Because we stacked ten O-rings, as mentioned above, the tilting angle range can be as wide as ±3.76◦. However, here we only show a smaller angle experimentally considering the phosphor screen size and the distance L. Figs. 5(a) and 5(b) present other experimental results in which the gun flange shifts horizontally, with

a shifting value of ±2 mm on the phosphor screen. This is nearly identical to the value obtained when using the clamping screws as a means of control. In order to differentiate between tilting and shifting motion, we adjust the only ∆V during the tilting while still maintaining the H1,2, and vice versa. When we loosen the clamping screws, the flange position returns to its original position because the O-rings have elastic restoring force. Therefore, the O-rings play a part not only in vacuum sealing but they also serve as a spring. To conclude, we successfully developed and verified an alignment method for an electron gun with an O-ring stacking structure. The direction of the beam can be tilted along two axes and the source position can be shifted in each of three orthogonal directions. We demonstrate a tilting angle of ±2.55◦ with shifts of ±2 mm experimentally, values which are generally sufficient for application to CPBI gun alignment, as this adjustable range can cover the error range when the parts are made and assembled. Furthermore, we can change the number of O-rings depending on the required range of adjustment and adapt to a conventional 6-in. flange type easily and quickly. Therefore, this is a good candidate for use in the early stages of CPBI development and in laboratory experiment which require low development costs. It can also be applied when insulation is needed between the flanges while maintaining some degree of freedom. This research was supported by Nano Material Technology Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (Grant No. 2011-0030233). 1J.

Hu and N. Tanaka, J. Electron Microsc. (Tokyo) 49(5), 651 (2000). Ohtsuki and L. W. Denney, U.S. patent 4,663,525 (5 May 1987). 3E. N. Beebe and V. O. Kostroun, Rev. Sci. Instrum. 63, 3399 (1992). 4T.-H. P. Chang, M. Mankos, L. P. Muray, H.-S. Kim, and K. Y. Lee, U.S. patent 6,288,401 B1 (11 September 2001). 5P. Adamec, E. Bauer, and B. Lencová, Rev. Sci. Instrum. 69, 3583 (1998). 2M.

Review of Scientific Instruments is copyrighted by AIP Publishing LLC (AIP). Reuse of AIP content is subject to the terms at: For more information, see

Note: O-ring stack system for electron gun alignment.

We present a reliable method for aligning an electron gun which consists of an electron source and lenses by controlling a stack of rubber O-rings in ...
4MB Sizes 2 Downloads 4 Views