Materials and Process Development for the High Volume Manufacturing of Complex Precision Devices
Expert Analysis
Developing a process and design for low cost, highly reproducible assemblies is always challenging. The Toyota Production Enterprise System, Six Sigma methods are increasingly being used to ferret-out waste. Equally important to the production of the device is the design and materials selection. Discussed below is work that was done at the in the Terrestrial Passives Design group at the Microelectronics Group of Lucent Technologies.
The three-pump combiner device shown in Fig. 1
The device consisted of three laser wavelengths which are combined onto one fiber (output collimator). The device, referred to as a three-pump combiner, uses light from each of the three collimators, pump 1, pump 2 and pump 3. The light is combined by designing two thin film dielectric band-pass filters to selectively transmit or reflect light from the three collimators. To aid in understanding the design complexity, a ray trace for pump 2 to the output collimator was imposed on Fig. 1 to illustrate the path of the laser light. It can be seen that the dielectric stack on both band-pass filters was designed to reflect light from pump 2. Light from pump 3 was transmitted through band-pass filter 2 and captured by the output collimator. Light from pump 1 was transmitted through band-pass 1 but reflected by band-pass 2 and then captured by the output collimator. The orientation of the collimators, relative to the filters, was critical to maintaining a low optical loss device. The requirement for environmental aging was that the loss be less than or equal to 0.5 dB. This was problematic since collimator angular misalignments of 0.4 degrees (i.e. tip and/or tilt) can result in an insertion loss (i.e. the optical loss of inserting an element in the optical path) of 1.34 dB.
Demonstration of device stability is a common requirement in many industries. For example, in the medical device industry, ISO 10993 is used to evaluate the biocompatibility of medical devices. In the telecommunications industry, all qualified passive optoelectronic devices must pass a series of rigorous tests detailed in the telecom standards Telcordia 1221 and 1209 standards. In brief, a statistical number, either 11 or 21, of manufactured devices must suffer less than a 0.5 dB insertion loss after a regimen of tests including shock, vibration, thermal cycling, and environmental aging. Depending on the number of starting devices for testing, either one or two failures are allowed. The insertion loss is simply the signal loss as a result of the insertion of an element in the optical path. The test that is most responsible for causing failure of adhesively mounted precision optics is the environmental aging test.
When adhesive materials are used to bond precision optics, failure of an optoelectronic device can occur by two mechanisms: loss of adhesion and swelling/dilatation. The latter may not result in failed adhesion; rather, the adhesive joint fails to hold the precision optics in position, resulting in unacceptable performance. Loss of adhesion is often characterized by an interfacial failure of the adhesive joint. Interfacial failure occurs when water displaces the van derWaals or hydrogen bonding functionality at the interface. Water can also serve to hydrolyze primary bonds, such as ester linkages, between the surface and the adhesive. The delamination of adhesive/glass interfaces is believed to occur by the displacement or hydrolysis of reactants at the polymer/glass interface.
Figure 2
Figure 3
Quantification environmental dimensional stability was demonstrated by pre-screening adhesive for their ability to maintain their shape after exposure to 3 days of boiling water. Because the extent of swelling is a function of the adhesive thickness, multiple adhesive bond lines were fabricated by bonding two glass slides together. The gap between the slide during bonding was controlled by precision slims. The distance between the shims was measured using the etalons created by the bonded specimens as shown in Figure 2. The output from this analysis is shown in Figure 3. Using this method the adhesive swelling could be easily quantified to 1 micron. This testing provided the basis of screen adhesive materials for dimensional stability.
To see the resume of the expert associated with this case study, see the link below.
