Raman spectroscopy – nuclearised

IT IS ESSENTIAL THAT THE nuclear industry is able to identify and characterise radioactive material and other hazardous substances. Facilities undergoing decommissioning have a great need for improved in-situ analytical capability so that fewer samples have to be transported for laboratory analysis. It is useful to have a wide range of sensors available to detect and identify as many potentially hazardous substances as possible.

Jacobs, the global technical solutions company, and IS-Instruments, a small tech business based in Kent in the UK which specialises in laser sensing technology, have developed a new class of spatial heterodyne Raman spectrometer and taken it from Technology Readiness Level 3 (TRL3) to TRL8. Mounted on a range of delivery platforms, including a robotic arm and a remotely operated vehicle (ROV), the Raman probe can identify key substances of interest to the nuclear industry from a distance of up to three metres.

How it works

Raman spectroscopy works by firing a low powered laser at the area of interest. The laser interacts with substances to produce characteristic light scattering known as a ‘Raman signal’. This signal produces a spectrum that is specific to each substance and can be used to create a chemical fingerprint to identify and quantify the materials present, even if there are many substances mixed together.

Classical radiometric tools such as gamma spectrometry can identify a wide range of radioactive materials. But they cannot easily identify alpha emitters and can struggle to detect low-energy gamma emitters in the presence of large quantities of other high-energy gamma emitters. Raman spectroscopy can identify radioactive materials and other chemical substances but because of the challenging nature of the nuclear environment, instruments often cannot be located close to the target substance. So, although Raman spectroscopy is a well-established and effective technology, it has been used very little in the nuclear industry.

Remote deployment

The ROV used was the Jacobs ROV022 high mobility platform, a nuclear-hardened tracked crawler with four independent drive pods that can be rotated allowing the system to climb over obstacles or alter its profile to fit within small spaces. Power and communication were delivered to the system via a trailed umbilical connection. These fibre couplings are used to enable all the sensitive electronics to be sited a considerable distance away from the deployment platform, enabling easier access to restrictive spaces for associated ROVs and reducing the damage caused by the hazardous environments found in nuclear facilities.

The ROV’s removable top plates provide a mounting point for a variety of different instruments, varying from manipulator arms to laser scanning systems. The probe was attached to a heavy-duty ‘pan and tilt’ mechanism fitted to the rear mounting point. A pan and tilt navigation camera mounted to the forward plate was used to locate and confirm the pointing direction for the instrument.

The Raman probe head was connected to the main spectrometer using a 15m long, 1mm diameter optical fibre. A shielded fibre was used to transmit the Raman signal from the probe head to the spectrometer. This followed initial testing where it was observed that moderate levels of ambient light were leaking into the unshielded fibre.

The probe head was developed with deployment into potentially radioactive environments in mind. Therefore, the external structure was designed to be easy to decontaminate. Ranging information was provided initially by a commercially available LIDAR system and allowed the focus control mechanism to be accurately positioned to target samples at various stand-off distances from the probe. Repeatability challenges later led to a stereo camera system being used to fulfil this function.

The compact and low cost technology was proven to be able to detect substances of interest to the nuclear industry in less than one minute, up to ranges of three metres. It uses the 3D stereo camera system to locate possible targets and provide ranging information to guide the focusing position of the Raman probe to maximise the signal.

Demonstration

Funding for the initial development was provided via the into Innovate UK’s ‘Game Changers’ competition. The system was further developed and incorporated into a wider decommissioning platform during Jacobs’ participation in the UK Integrated Innovation for Nuclear Decommissioning (IIND) competition, which was supported by Innovate UK, the Nuclear Decommissioning Authority and Sellafield Ltd.

The system has been tested in a representative environment. Observations were made of substances including cellulose, uranyl nitrate, mixed organic liquids (tri- butyl phosphate and kerosene) and a thin smear of tri-butyl phosphate on a stainless steel surface.

During the full scale demonstration phase of the IIND project, the technology displayed its flexibility by being deployed via a robotic arm in a full scale mock-up of a Sellafield fuel separations cell where it successfully confirmed its ability to scan the internals of the cell, and using bespoke machine learning algorithm software, detect and accurately identify a range of substances of interest to the nuclear industry.

The feedback from the panel of expert assessors was that the characterisation technologies, and in particular the Raman technology, was of high maturity. It demonstrated clearly the importance and usefulness of high quality in-cell characterisation in decommissioning planning and operations.

Jacobs was chosen as the priority project team to deploy the technology in decommissioning an active separation cell on the Sellafield site. The Raman technology will be one of the first to be deployed.

The performance so far of this simple, low cost and robust technology shows that it can help in solving a number of complex nuclear industry characterisation challenges.

Author information: Mike Wharton, Senior chemist and technical lead, Jacobs; Jon Storey, Director, IS-Instruments