New developments in advanced welding

Micro-electron beam welding

Owing to the very high functional integration density of the components and also to the great variety of materials, micro-systems and, in particular, hybrid micro-systems are making high demands on joining techniques. For example, the minute dimensions of the components and the frequent joining of dissimilar materials with differing thermophysical properties require joining methods that convey their energy selectively and locally in a minimum of space. Both laser beams and electron beams are suitable tools for this purpose. Moreover, processing in high vacuum meets the demands of a high-purity environment.

7.4.1 Technology

Electron beam welding machines from the macro-range cannot be used for micro-components. This is because their beam powers lie between 100 W and several kW which are too high for the welding of microfine components. The use of a scanning electron microscope (SEM) as a welding tool seems much more promising. This type of machine combines two basic functions, observation and welding, in just one piece of equipment. Calculations show that a maximum acceleration voltage of 30kV and a probe current of 200 mA give a maximum beam power of 6W in the beam generator. Power losses caused by screening through a diaphragm and scattering in the beam column reduce the power to approximately 3 to 4W at the workpiece.

Figure 7.11 shows the beam path of the welding equipment in both operating conditions, observation and welding. The integration of both functions into one piece of equipment makes opposing demands on the technique. The observation and analysis of substrates require low energy input and high resolution. The high number of diaphragms that are small enough to screen edge electrons and the two condenser coils that reduce superfluous electrons in the beam cause extremely small beam radii. On the other hand, in most welding applications a higher power, by several orders of magnitude, is necessary. The technical modifications that have been carried out successfully are reversible and are primarily concerned with the electromagnetic components. Among such modifications are the removal of two diaphragms from the liner tube and the increase of the aperture diaphragm diameter as

well as the switching off of the first condenser coil. Because these modifications are reversible, the existing equipment may be used both for observation and for welding (Carslaw and Jaeger, 1967).

In the practical applications of the micro-system technique, micro components are joined to one another in varying arrangements. The quality of a welded joint is strongly dependent on the adjustment precision of the components to be joined, among other factors such as joint preparation. For instance, a slight angle deviation of two components, joined with a square butt joint can lead to a gap no longer compensated by the minute beam diameter of a few micrometers. Faulty joints are a consequence. Owing to the excellent observation possibilities of SEM, the highly precise adjustment is carried out only in its working chamber. This means that the existing coordinate system in the interior of the compound chamber has, with the five-axes macro table, been completed to a second independent coordinate system. In addition, a subsequent correction of the component position without repeated withdrawal of the components from the compound chamber is possible.

The adjustment device is composed of three linear adjusters, two tilting axles and one control unit. The vacuum suitability of the components allows the maintenance of the necessary vacuum and the reliable operation of the motors. Self-locking of the gears allows the controls to be switched off after reaching the desired position and leaves the electron beam uninfluenced by electromagnetic fields.

When choosing and integrating the second adjustment unit into the vacuum chamber of the SEM, several conditions must be considered:

• Vacuum suitability of the mechanical and the driving components;

• High plane-parallelism of adapter plates;

• Circumvention of collision with wall, electron gun and detectors during

operation;

• Centre position of all axes below the exit outlet of the electron beam;

Carrier

3 linear stages and 1 rotating stage (SEM),

X-, Y-, Z-axes, 0Z

7.12 Diagram of partitioning unit.

• Z-position of the joining plane must be located in the region of the working distance of the electron beam;

• Vacuum pipe for the electrical drive of the motors.

The system design allows one joining component to be moved by the five-axes-positioner independently of the second joining component. Figure 7.12 shows the positioning unit and its diagrammatic representation after installation in the SEM.

7.4.2 Micro-welding process

Process sequence

Figure 7.13 shows the chronological sequence of the welding process. As a first step, the components to be joined are adjusted exactly in relation to each other and the electron beam is positioned on the joint. There is then a changeover to the welding mode so that the actual welding operation can start; however, on-line process observation is not yet possible. After the welding process is completed, the joining point may, after changeover to the observation mode, be subject to further analyses or measurements. In practice all welding sequences correspond with this step sequence.

Process variations

In practice, there is the choice of several welding methods, (Fig. 7.14). They basically differ in the type of beam manipulation on the substrate. During single scanning the electron beam is guided once over the welding zone with

• Evaluation of the position

a fixed welding speed; during multiple scanning it oscillates for a certain period of time with a preselected deflection frequency. The choice of method depends on the joining task. Single scanning for the joining of foils leads to very clean, sharply delimitated weld edges without weld notches; multiple scanning, however, shows weld edge regions that are remarkably elaborated. The beam energy, absorbed over a longer period of time, leads to the partial evaporation of the metal. Energy input can be varied further by the scanning of a larger substrate surface on a lateral level. Here the electron beam is applied as the heat source for a soldering process. In a first step, low-melting soldering materials are applied on the micro-components that are to be joined. The electron beam leads, through its thermal influence, to the development of intermediate constituent phases inside the soldering materials and these in

turn result in the joining of the components. In particular, materials with good thermal conductivity, or non-metals may, under certain prerequisites, be joined by means of micro-electron beam soldering.

7.4.3 Examples of joining

Knowledge about the methodology and equipment of electron beam welding has primarily been applied to wire joints and metal foils. Additionally, weld - in tests have been carried out to investigate the behaviour of silicon and plastics. Joining thermoelements made from NiCr-Ni wire combinations with a wire of diameter 70 mm each allows almost globular beads for temperature measurement in the micro range. In the field of plastics, polyethylene showed, after prior gold plating, good welding results. By means of materials with favourable heat-conductive properties, such as copper or aluminium, micro-soldering with the electron beam as the heat source is examined using Cu-Sn soldering systems (Janssen, 1991).

7.4.4 Summary

The understanding of the processes concerning electron beam welding in micro-technology and the extension of the equipment technique are consolidated by further research work in the fields of beam characterisation, beam-material interactions, temperature measurements and the integration of image processing in the control of welding processes. An important future aspect of research is the investigation of welding tasks with optional weld seam geometries. The production of appropriate control for the beam deflection and of means of suitable heat control are the most important challenges.