Sensors for process control of personal care products
Lead Research Organisation:
University of Manchester
Department Name: Chem Eng and Analytical Science
Abstract
The main research question is to research and develop a method for measuring in line viscosity of non-Newtonian complex fluids with particular applications to personal care products such as hair shampoos and conditioners. The objective is to develop and test a appropriate methods in the lab in static samples and then build up to demonstrating it in Unilever's pilot plant facility and/or manufacturing plants. The motivation for addressing this problem is that currently samples being processed have to be extracted and taken to a nearby laboratory for manual measurement of viscosity. With the returned result, changes or further processing may need to be made to the produced product and the overall process. If an in-line sensors could be developed then there is the scope for significant wastage reduction, energy reduction and time spent in processing leading to increased efficiency. A smart sensor would also contribute to the overall process control of the system.
The approach being investigated is to explore non-invasive optical spectroscopic techniques such as mid-infrared and near-infrared spectroscopy and Raman spectroscopy. Suitable probe designs will also need to be investigated. The data will require advanced multivariate data analytical techniques to correlate any changes in the spectra with viscosity. Studies of different feedstocks and different process regimes such as flow rates will be considered in addition to the influence of temperature which strongly influences viscosity. The complication is that the types fluids under consideration are complex and behave in a non-Newtonian manner.
Fundamental understanding of the physical origin of any correlations will be explored in terms of changes to the microstructure of the materials and how this influences viscosity.
The research fits into the EPSRC Future Manufacturing and digital manufacturing themes.
The approach being investigated is to explore non-invasive optical spectroscopic techniques such as mid-infrared and near-infrared spectroscopy and Raman spectroscopy. Suitable probe designs will also need to be investigated. The data will require advanced multivariate data analytical techniques to correlate any changes in the spectra with viscosity. Studies of different feedstocks and different process regimes such as flow rates will be considered in addition to the influence of temperature which strongly influences viscosity. The complication is that the types fluids under consideration are complex and behave in a non-Newtonian manner.
Fundamental understanding of the physical origin of any correlations will be explored in terms of changes to the microstructure of the materials and how this influences viscosity.
The research fits into the EPSRC Future Manufacturing and digital manufacturing themes.
People |
ORCID iD |
| Philip A Martin (Primary Supervisor) | |
| Kiran Haroon (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/R512035/1 | 01/10/2017 | 31/12/2022 | |||
| 1920513 | Studentship | EP/R512035/1 | 01/10/2017 | 30/09/2021 | Kiran Haroon |
| Description | It has been found that spectroscopic techniques (near infrared (NIR), mid infrared (MIR) and Raman) combined with machine learning techniques (mainly partial least squares regression) can be used to predict the viscosity of shampoo inline both as individual techniques and combining datasets. Outcomes: Haroon, K., Arafeh, A., Martin, P., Rodgers, T., Mendoza, C. and Baker, M. (2019), Use of inline nearinfrared spectroscopy to predict the viscosity of shampoo using multivariate analysis. Int J Cosmet Sci, 41: 346-356. Haroon, K., Arafeh, A., Cunliffe, S., Martin, P., Rodgers, T., Mendoza, C. and Baker, M. Comparison of individual and integrated inline Raman, near-infrared and mid-infrared spectroscopic models to predict the viscosity of micellar liquids (Under review - Applied Spectroscopy) |
| Exploitation Route | Currently, there are inline viscometers available commercially, however they have been shown to have issues related to data acquisition times, representative measurements, robustness and cleanliness. The studies here show that using commercially available inline spectroscopic techniques (inline immersion probes) combined with partial least squares regression to predict the viscosity of a complex non-Newtonian fluid can be achieved. Although this work focuses on personal care products, viscosity models could be developed for many other products with the intent to eliminate the need for laborious offline testing. Having the means to monitor viscosity throughout the process would lead to better process understanding, more consistent products of high quality and a more efficient process with the potential to use this measurement as a means of process control. |
| Sectors | Manufacturing, including Industrial Biotechology |