Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Agricultural and Biological Engineering

Committee Chair

Osvaldo H. Campanella

Committee Co-Chair

Bradley L. Reuhs

Committee Member 1

Carlos M. Corvalan

Committee Member 2

Owen G. Jones

Committee Member 3

Ganesan Narsimhan


Understanding the relationship between structure and functional properties in plant-cell-wall-derived foods has become a growing interest to both academia and industry. Tomato is one of the most cultivated vegetable crops and mostly is consumed as processed products in the form of suspensions. Rheological properties of tomato product, a key functional attribute, depends on both the serum and particle phases of these products. Although recent studies have suggested that the particle phase is the dominant factor, the relationship between fundamental particle properties and the bulk rheology of the suspension is still unclear. This research systematically evaluated the contributions of soluble pectin and particle phase on the rheology of tomato suspensions, and identified that the particle structure and its physical properties are crucial in determining the rheology of such systems. Alteration of these properties either by processing conditions or by internal enzymatic activity could cause a significant change in the rheology of tomato products.

The serum phase of the suspensions displayed a Newtonian behavior with a low viscosity (~0.1 mPa.s). The contribution of soluble pectin to the overall viscosity of the suspensions was found to have a little influence despite that reconstituted suspensions were prepared either with large pectin concentrations or with pectin having a high degree of methylation. However, the presence of pectin was important because its role on stabilization of the suspension systems by increasing the interaction between particles. When pectin concentration was low, wall slippage during measurements was observed due to phase separation by using cone-plate geometry. A vane geometry was able to alleviate the slippage artifact and a good correlation (R2=0.91) was found between the empirical Bostwick consistometer method and fundamental measurements performed employing the vane geometry. Hence, the vane geometry was recommended in the viscosity measurements of cell-wall-derived suspensions.

The particle structure and its physical properties, and the associated particle interaction controlled the rheological properties of the cell-wall-derived suspensions. Changes in the particle phase were achieved in this study by two means: external processing with various conditions and molecular biological modification by reduced pectin methylesterase (PME) activity. The effects of thermal breaking, and physical treatments such as ultrasound and high shear were employed at the laboratory scale. The concentration process to produce tomato paste from tomato juice at an industrial scale was also investigated. The focus was on effects that this process has on the properties of the particles and the rheology of the suspensions when they are reconstituted from the paste to juices. These diverse processing and modification conditions produced particles with various structures and strengths, and as a result caused significantly changes on the rheological properties of suspensions.

Although both the ultrasound and high shear treatments reduced significantly the particle size of the treated tomato suspensions, the former led to an increase in their rheological properties whereas the latter caused a significant decrease. It could be explained by formation of particles with structural differences provoked by these two treatments. Ultrasound treated suspensions contained more intact particles, and with large strength, which was evaluated by a compression test on a limited number of particles. Conversely, high shear treated suspensions resulted in mostly ruptured particles that lost mechanical strength. The water-soluble pectin (WSP) fraction increased after ultrasound and shear treatments. However, soluble pectin is not the direct cause for the changes in the suspension rheology; it is an indicator or consequence of the changes in particle properties.

This research also explained the viscosity loss during the industrial tomato juice concentration process from the perspective of particle alterations. The particle phase was extensively modified as the concentration process reduced the particle volume and concentrated its mass into a smaller size. The original tomato juice had a relatively higher volume fraction and viscoelasticity than those of reconstituted juices from dilution of pastes to achieve the same soluble solids (oBrix). This resulted in original juices with higher consistency and viscosity. During dilution, paste particles cannot re-expand to the original shape and volume than those present in the original juice. Due to the fact that the concentrated solute present in pastes cannot be fully solubilized, more paste is necessary to achieve the viscosity of the original juice.

In addition, tissue structure modification using molecular biology and via suppression of pectin methylesterase (PME) activity resulted in a closely packed cellular structure with smaller pore size when compared to the tissue of the original wild type tomato (OWT). An 85-90% reduction in PME activity significantly strengthened the microstructures of cell wall particles, and reduced serum separation, which improved tomato suspension rheological properties.

The last part of this research investigated the flow behaviors of industrially processed hot-break (HB) and cold-break (CB) tomato suspensions under steady-state and dynamic oscillatory shear conditions. The HB suspensions exhibited considerably higher viscosity and viscoelastic properties than CB suspensions because their particles had a structure that was able to retain better water and higher mechanical strength. Both industrially processed samples exhibited temperature-dependent and time-dependent rheological behaviors. The consistency coefficient (k) as a function of temperature could be modeled by an Arrhenius-like equation. The activation energy of the HB sample was higher than that of the CB sample, indicating a more integral structure resisting changes in temperatures. The thixotropic behavior of HB and CB suspensions was described by the Stretch Exponential equation. A characteristic time ( s  ) used in the Stretch Exponential equation increased with temperature for the HB sample whereas it showed the opposite trend for the CB sample. These differences could be explained by differences in the particle structure and initial viscosity. Particle interactions showed great impact on the rheological properties. When particle concentration was low (solid % < 1.0%), both HB and CB samples almost had the same apparent viscosity due to a limited contact between particles. However, when the particle phase was high, the particle-particle contact significantly increased, and the HB sample demonstrated a considerably higher viscosity and viscoelasticity. Results indicated that the HB system has larger particle elasticity and stronger particle interaction than the CB system. Furthermore, the local Young’s modulus distributions of individual HB and CB particles investigated by Atomic Force Microscopy (AFM) were in good agreement with the bulk rheology data. It can be concluded that the differences in rheological properties of tomato products are originated from differences in their particle phases.