Opendata, web and dolomites


Teaser, summary, work performed and final results

Periodic Reporting for period 1 - REDFINCH (mid infraREd Fully Integrated CHemical sensors)


REDFINCH – Compact, portable chemical sensors.The REDFINCH project is developing the next generation of miniaturised, portable optical sensors for chemical detection in both gases and liquids. Initial target applications are in the petrochemical and dairy industries.REDFINCH...


REDFINCH – Compact, portable chemical sensors.

The REDFINCH project is developing the next generation of miniaturised, portable optical sensors for chemical detection in both gases and liquids. Initial target applications are in the petrochemical and dairy industries.
REDFINCH is a consortium of 8 European research institutes and companies, undertaking cutting edge R&D into novel, high performance, cost effective chemical sensors, based on Mid-Infrared Photonic Integrated Circuits (MIR PICs). Silicon PICs — integrating optical circuits onto millimetre-size silicon chips — create extremely robust miniature systems, in which discrete components are replaced by on-chip equivalents. This gives a simultaneous improvement in ease of use and a reduction in cost.

The REDFINCH partners are addressing the significant challenge of implementing this in the important mid-infrared region (2-20 μm wavelength range), where many important chemical and biological species have strong absorption fingerprints. This allows for the detection and concentration measurement of a wide range of gases, liquids and biomolecules, crucial for applications such as health monitoring and diagnosis, detection of biological compounds, monitoring of toxic gases, and many more. Initially, REDFINCH will focus on three specific key applications:
- Process gas analysis in refineries
- Gas leak detection in petrochemical plants and pipelines
- Protein analysis in liquids for the dairy industry

Work performed

The REDFINCH project started in December 2017. Refer to the project website,, for updates on work performed and results as the project progresses.

First, regarding the integration technologies on Si, homogeneous and heterogeneous QCL and ICL sources are beginning to stabilize. The main objective is to lower the cost of these sources in order to make their intrinsic qualities widely available. Basically, one can rely the cost of MIR lasers to the costs related to 1) epitaxy, 2) manufacturing yield, 3) packaging. It can be seen that the project work presented in the technical periodic report contributes to reducing these 3 items. The project objectives in for these components therefore remain credible. A precise costing will have to be made in the coming months, but a reduction by a factor of 10 seems achievable in the short term.
In addition, the first photoacoustic tests with a cell of a few dozen mm² show performances compatible (i.e. below ppm) with applications requiring small footprint and high performance such as a leak detector. This component could be used as is, as in the mirSense multisense module. In addition, it opens the way to even more integrated versions such as µPa.
Regarding the photonic circuits, different platforms have been investigated to get the relevant index constrast for each application : SiGe/Si, SiGe/SiN, Si/SiO2. Finally, the on-chip protein sensor is now defined and currently being fabricated on Si 200mm pilot line. For 3-4µm PIC wafers are now been fully fabricated and include a variety of gratings, ring restonators and photonic crystals, which will be used in the first instance, to realise hybrid lasers.
Finally, Redfinch website and conference communication have made it possible to establish preliminary contacts with several industrials.

Final results

Despite the importance of the mid-infrared (MIR) wavelength region for a wide range of application areas, current state-of-the-art sensing systems in the MIR tend to be large and delicate. This greatly hampers their deployment in real-world applications.
REDFINCH will harness the power of Photonic Integrated Circuits (PICs), using hybrid and monolithic integration of III-V diode and Interband Cascade/Quantum Cascade materials with silicon to create high performance, cost effective sensors. Integration creates extremely robust systems, in which discrete components are replaced by on-chip equivalents, giving a simultaneous improvement in ease of use and a reduction in cost. Silicon photonics leverages the advantages of high performance CMOS technology, providing low cost mass manufacture, high fidelity reproduction of designs, and access to high refractive index contrasts that enable high performance nanophotonics.
REDFINCH will realise three fully-integrated PIC-based chemical sensor demonstrators for:

Refinery gas is released during the fractional distillation of crude oil into various products of commercial use, such as butane, diesel and gasoline amongst others. Refinery gas is a mixture of multiple gases, consisting mainly of non-condensed hydrocarbon gases, C1 to C5. As the composition of refinery gas varies widely depending on the origin of crude oil, large-scale refining operations require fast decisions as the parameters of the process change.
- The continuous monitoring and fast response time enabled by the REDFINCH sensor will greatly enhance process efficiency.

The consumption of natural gas is steadily increasing worldwide; in Europe the construction of new pipelines and ports to deliver liquefied natural gas (LNG) from different sources to the European gas grid will establish reliable gas supply at low cost for the next decades. This gas grid will also serve as a huge storage system to compensate for the increasing share of fluctuating renewable energies in Europe. The infrastructure needs continuous monitoring, especially as it ages. Manual inspection, mainly using handheld portable sensors, requires a high amount of personnel and is very costly and time consuming.
Photonic Integrated Circuits with micro Photo-Acoustic Sensors (PIC-μPAS) developed in REDFINCH will be a key innovation for leak detection. These sensors will have a consumption of less than 10 W in continuous operation. For infrastructure monitoring, they will be operated in a slow pulse-burst mode. If a leak is detected the awareness of the sensor, i.e. the pulse frequency will be increased. By such methods, average power consumption is very low and the sensor could be battery-operated over a year or powered by an ambient energy harvester e.g. a solar cell. PIC-μPAS will be part of a wireless sensor network to monitor e.g. a refinery infrastructure.
- Compared to existing sensors, they will be more sensitive, more specific and of higher dynamic range resulting in more reliable detection and less false alarms.

Milk production and processing is a multi-billion Euro industry. Farmers want to monitor the milk to better understand the health of their cattle and manage the herd. Dairies need to monitor the milk as the farmers are paid according to the constituents of the milk. Processors need to monitor the milk to quality control their products. Today nearly all milk monitoring is done offline through the taking of samples and the analysis of those in a lab – a slow, expensive process. State of the art milk analysers cannot differentiate between different types of proteins, they can only quantify the total protein content in a sample.
- The REDFINCH protein sensor will enable analysis of the milk at the collection tanker, allowing efficient redirection and distribution of the consignment.

Website & more info

More info: