Breakthrough in the development of stretchable optical waveguides
A range of emerging applications requires electronics to dynamically adapt to curving and bending surfaces. Wearable on-body sensing systems or instruments in a cockpit integrated on curved surfaces are just two examples being investigated, and others can be found in applications domains like robotics and automation, healthcare and biomedical technologies, and consumer electronics.
In recent years, progress has been reported on the development of stretchable electrical technologies, in which, for example, non-stretchable islands are embedded in a stretchable material and joined with stretchable electrical interconnections.
Belgian electronics research institute imec, in partnership with CMST, imec’s associated lab at Ghent Univ., have recently transitioned stretchable technology into the optical realm. By integrating the light sources and detectors, these teams have fabricated the world’s first randomly deformable optical waveguide. It remains functional for bending radii down to 7 mm, and can be stretched to more than a third of its length. These optical links can be used to interconnect optical components within a stretchable system, just like stretchable electrical interconnections are used for stretchable electrical systems.
More and more sensing systems, such as artificial skin, are implemented using optical instead of electrical technologies. For these applications, stretchable optical interconnections are needed instead of electrical interconnections. When compared to electrical interconnections, optical interconnections are much more difficult to make, for two main reasons. First, their dimensions are much smaller, typically a few micrometers up to tens of micrometers in cross section. And, secondly, while for stretchable electrical connections the alignment is not that critical, the accuracy needed for an efficient coupling between waveguides and light sources/detectors is much higher.
Imec’s approach to making stretchable optical interconnections is based on multimode PDMS (poly-dimethylsiloxane) waveguides. Such a waveguide consists of an optically transparent channel of a stretchable PDMS material (called the core of the waveguide), surrounded by another type of PDMS material (called the cladding of the waveguide) with a lower refractive index. Due to this specific configuration, light is trapped in the channel and propagates along its length. This will remain valid when the waveguide is deformed. The team selected two commercially available PDMS materials with a precise refractive index difference for core and cladding. The waveguides, having a cross-section of 50 µm x 50 µm, were patterned by using a replication technology based on the capillary filling of the PDMS microchannels.
To see how far the waveguides could bend and stretch before too much lights escapes, a process was developed to integrate light sources (VCSELs with ʎ = 850 nm) and detectors (GaAs based photodiode) with the waveguide. This way, the researchers obtained a truly bendable and stretchable optical link. They characterized the waveguides in terms of propagation losses, bending losses, stretching losses and reliability.
For bending radii down to 7 mm—which is about the diameter of a human finger—only small losses were observed. And the link remained functional during stretching up to 30% elongation (observed extra loss below 0.7dB). Also, the waveguides showed high reliability, as they did not see any degradation in the material even after mechanically stretching it 80,000 times at 10% elongation. They did however observe some additional loss fluctuations during stretching, which can be ascribed to the changing coupling conditions at the input and output regions of the waveguides, i.e., the parts where light enters and exits.
Imec’s team believes this is the first truly bendable, stretchable optical link. Since commercially available materials have been used and a replication method was used for patterning the waveguide, the proposed technology is widely applicable and cost-efficient. The main future work will be focusing on making the waveguides smaller, and on finding appropriate ways to integrate them in specific applications. Also, the mechanical stability at the interfaces with the optoelectronic components needs to be further improved to minimize additional loss fluctuations.