Volume 57, Issue 9 p. 686-694
SPECIAL ISSUE RESEARCH ARTICLE

Early stages of fat crystallisation evaluated by low-field NMR and small-angle X-ray scattering

Marjorie Ladd Parada

Corresponding Author

Marjorie Ladd Parada

School of Food Science and Nutrition, University of Leeds, Leeds, UK

Correspondence

Department of Physics, AlbaNova University Center, Stockholm University, SE-10691 Stockholm, Sweden.

Email: [email protected]

Search for more papers by this author
Megan J. Povey

Megan J. Povey

School of Food Science and Nutrition, University of Leeds, Leeds, UK

Search for more papers by this author
Josélio Vieira

Josélio Vieira

Product Technology Centre, Nestlé, York, UK

Search for more papers by this author
Michael Rappolt

Michael Rappolt

School of Food Science and Nutrition, University of Leeds, Leeds, UK

Search for more papers by this author
Michael E. Ries

Michael E. Ries

School of Physics and Astronomy, University of Leeds, Leeds, UK

Search for more papers by this author
First published: 06 March 2019
Citations: 10

Abstract

Low-field time-domain nuclear magnetic resonance (NMR; 20 MHz) is commonly used in the studies of fats in the form of solid fat content (SFC) measurements. However, it has the disadvantage of low sensitivity to small amounts of crystalline material (0.5%), thus often incorrectly determining crystallisation induction times. From spin–lattice relaxation rate measurements (R1) during the isothermal crystallisation measurements of cocoa butter between 0.01 and 10 MHz using fast field cycling NMR, we learnt previously that the most sensitive frequency region is below 1 MHz. Thus, we focused on analysing our 10-kHz data in detail, by observing the time dependence of R1 and comparing it with standard SFCNMR and SFC determinations from small-angle X-ray scattering (SFCSAXS). Although not reflecting directly the SFC, the R1 at this low frequency is very sensitive to changes in molecular aggregation and hence potentially serving as an alternative for determination of crystallisation induction times. Alongside R1, we also show that SFCSAXS is more sensitive to early stages of crystallisation, that is, standard SFCNMR determinations become more relevant when crystal growth starts to dominate the crystallisation process but fail to pick up earlier crystallisation steps. This paper thus demonstrates the potential of studying triacylglycerols at frequencies below 1 MHz for obtaining further understanding of the early crystallisation stages of fats and presents an alternative and complementary method to estimate SFC by SAXS.