Optical sensors can measure ingredients in foods
Nofima research fellow Tiril Lintvedt has investigated whether it is possible to carry out detailed quality measurements of raw materials directly on the production line. Her research shows that measurements with Raman spectroscopy offer many possibilities.
More specifically, Tiril has investigated whether it is possible to use Raman spectroscopy to measure fatty acid composition in salmon fillets or fat, protein, bone and collagen in residual raw materials from chicken and turkey.
Technological development opens up new opportunities
The technological development gives the food industry far better opportunities for process control and product differentiation, quality control and sorting of raw materials and raw materials. Optical sensors, used in spectroscopic measurement methods, are becoming faster, less expensive and more precise.
There are several different rapid and non-destructive spectroscopic measurement methods. They are suitable for different raw materials and measurement needs.
Detailed measurements directly on the production line are challenging
“Raman technology, which I have studied, has become both more affordable, better and more stable. It is now possible to make exact measurements over larger areas than before, but there are still challenges associated with measuring at exposure times low enough to keep up with the speed of, for example, conveyor belts. In addition, the measurements are affected by large variations in working distance as a result of varying thickness and volume of raw materials,” says Tiril Lintvedt.
The challenges she mentions relate to how long exposure time is needed to capture the necessary details in the raw materials. If one is to measure, for example, salmon fillets while they move on the production line in normal production, the measuring instrument that in Tiril’s experiment is a Raman probe must measure both very detailed and precise – and quickly.
“There are as far as we know no inline Raman solutions in the food industry today. Current spectroscopy solutions in use, such as NIR, are less sensitive to variations in distance or thickness, but current solutions fail to measure with the same richness of detail as Raman. For now, therefore, Raman measurements are better suited for rapid measurements alongside the production line, for single samples. Such measurements can also be of high value to the industry. They provide opportunities for frequent feedback on quality, and as of today there are no such solutions,” Tiril points out.
From laboratories to industrial production
Very detailed measurements are needed, right down to the molecular level, to map fatty acid composition in salmon fillets or fat, protein, bone and collagen in residual raw materials from chicken and turkey. Raman spectroscopy is particularly suitable for such measurements.
“Previously, Raman was used in laboratories, in the development of expensive products such as medicines. Better and cheaper solutions open new opportunities for industrial use also in food production. It is this potential that I have explored in my PhD, and the results show great potential for the development of new Raman applications in quality documentation, sorting, process analysis and real-time process control in the food industry,” says Tiril.
After completing her PhD, Tiril continues as a postdoc in DigiFoods – Centre for Research-driven Innovation, and will continue the work of developing Raman applications for the food industry.
Facts about the doctoral degree
Tiril Lintvedt defends her thesis on 11 May. The title of the thesis is Raman spectroscopy for in-line food quality characterization.
The public defence will take place at Nofima in Anton Skulberg’s Auditorium. The trial lecture starts at 10.15. Supervisors are Professor Achim Kohler (NMBU), senior scientists Jens Petter Wold and Nils Kristian Afseth (both at Nofima).
The doctoral degree is financed by Nofima and the Research Council of Norway. Current projects are, respectively, Spectec, SFI DigiFoods and Matpiloten (grant numbers 309259, 296083).