Development of a Process Approach for Retaining Seaweed (Sugar Kelp) Nutrients
Project Description
S. latisimma, commonly known as sugar kelp, belongs to the Pheophycae class of marine macroalgae. Freshly harvested kelp is either sun-dried or hot-air dried to extend its shelf-life by removing excess moisture. High temperatures and long exposure time during drying induces deteriorative chemical changes resulting in significant loss of bioactive compounds such as fucoxanthin and phlorotannin, which are not present in terrestrial plants. The aim of this study was to develop a simultaneous heat and mass transfer model in COMSOL using the thermophysical properties (thermal conductivity (k), specific heat capacity (Cp), thermal diffusivity (D), bulk density (ρ), porosity (ϕ)) of kelp calculated using Artificial Neural Networks (ANNs) to predict the energy requirement and drying kinetics under applied drying conditions.
Results and Accomplishments
Seaweed samples were obtained from Maine Fresh Sea Farms (Bristol, ME). Samples with moisture contents (MC) in the range of 0.05 – 9.0 (dry basis) were prepared by rehydrating freeze-dried kelp powder. The ρ and ϕ of the samples was estimated using displacement method and mercury intrusion porosimeter, respectively. Six replicates of k, Cp and D values were measured using KD2 pro analyzer at different temperatures (T) (30-70°C) and were fitted in empirical equation considering the individual and interaction factor (MC x T). The ANN model evaluated the different configurations of hidden layers and neurons for 180 data points in MATLAB. The 3D AutoCAD model of dryer and kelp was solved for heat and moisture transport along with airflow modeled using RANS k–ε turbulent equation in COMSOL.
The ρ values increased from 777-940 kg/m3 with MC, whereas k, Cp and D varied with MC and T and were in the range of 0.147-0.546 W/m.K, 0.605-2.727 MJ/m3.K and 0.243-0.313 mm2/s, respectively. Statistically, the linear (MC and T), quadratic (T2) and MC x T factors were significant (p < 0.05). Preliminary results showed the ANN model consisting of two hidden layers and six neurons was able to predict the thermophysical properties more accurately than empirical equation with a relative mean error of less than 15%.
This project represents the first study to report the thermophysical properties of sugar kelp or any seaweed. The developed model will help in optimizing the design parameters of large-scale dryers to obtain uniform product quality. Project researchers secured federal funding from NOAA for an integrated project on sustainable processing of kelp. With the new project and support from SEANET, researchers are reaching out a broader section of stakeholder for sustainable processing of sugar kelp. Results will help seaweed producers and processors in the region better prepare for producing high quality seaweed products. The developed designs and techniques will reduce the energy cost and help process the crop more efficiently so stakeholders may become economically competitive in the US and abroad.
Summary of Data Being Collected
| Data | Type | Quantity | Location |
| Proximate analysis | Quantitative | pH, Ash content, color | Food Science Laboratories, UMaine/Maine Fresh Sea Farms cultivation site |
| Antioxidant capacity, vitamin C, color | Antioxidant capacity using DPPH assay, Mass concentration and color intensity | Samples run at 3 or 4 different humidity levels, three different temperature levels | Food Science Laboratories, UMaine |