Fri, 24 September, 2021
In the SEER project, process monitoring sensors are being developed that are based on very small silicon-based optical chips, which look similar to the electronic chips used in our computers and smartphones. The difference is that they house optical circuits instead of electronic circuits for their operation.
These optical chips, sometimes-called ‘photonic integrated circuits (PICs),’ have recently gained a lot of attention as they allow the miniaturisation of very promising and highly sensitive accurate sensors into a compact housing.
The sensors use light instead of electricity for their operation, which means we need to get light into the chip for the sensor to operate.
In SEER, we will use the sensors to measure temperature, pressure and refractive index during the manufacturing of composites. We will send specific light into the sensor structures on the chip, and those structures will interact with temperature, pressure and refractive index in such a way that the properties of the input light are changed in a predictable way. Because of this, we also need to get the light out of the chip so we can evaluate the changes and link them to changes in temperature, pressure and refractive index.
The chip has several test structures that allow development and evaluation of the sensor structures.
Why is it a challenge to get light in and out of optical chips?
It is a challenge to get light in and out of the optical chip because the structures are so small. For example, the waveguides, which are transparent channels through which the light is guided on the chip, have cross-sectional dimensions much smaller than 1μm x 1μm.
For this reason, grating couplers have been invented to change the direction of the light, so that light which is being guided on the chip, can be coupled out nearly vertically with respect to the surface. This also works in reverse as external light can be launched onto these grating couplers and will then be guided by the waveguides on the chip.
The grating couplers are also designed to perform a conversion between the very small on-chip structures and the larger optical fibres that are commonly used to get light in and out of the optical chips.
The optical fibres are similar to the optical fibres used for high-speed telecommunication over large distances. They are essentially also waveguides for the light and, although their diameter of 10μm makes them much larger than the on-chip waveguides, they are still very small.
How do you get the light in and out of these small optical sensor chips?
Usually, an optical fibre, or multiple optical fibres are directly attached to the optical chip to form a permanent connection through which the light can get in and out of the chip. This requires a very precise alignment to make sure the light is coupled efficiently.
This method is, for example, a common practice in telecoms applications. For sensing, however, there is a problem because there are sensing structures at the top surface of the chips and, if a fibre was attached at the top, it would block or hinder the active sensing areas.
Working as part of the project consortium, IMEC are developing a new method that allows light to be coupled to the chip from behind rather than from on top. This ensures the sensor structures can operate without any obstruction as well as allowing efficient packaging of the sensor chip in a housing with cross-sectional dimensions, which can be as small as the dimensions of the chip itself.
One of the solutions investigated employs microlenses that are fabricated on the back of the optical chip to ensure that light from a fibre, which is placed below the chip, is precisely focused onto the top side of the chip. Without the microlens, the distance between the fibre and the chip’s top side would be too large, because the chip has a thickness of about 0.7mm. The microlens-based approach was previously demonstrated by IMEC for mid-infrared sensing and will now be further developed for the SEER sensors.
Another solution being investigated employs a ball lens that is placed between the bottom side of the optical chip and an optical fibre, which is placed below. The difference with this solution is that this ball lens is a separate component, which needs to be precisely aligned with both the chip and the fibre, meaning that precisely fabricated holders can be employed.
The SEER project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871875.