Alginate monomer composition studied by solution- and solid-state NMR – A comparative chemometric study Tina Salomonsen a, b, * , Henrik Max Jensen b , Flemming Hofmann Larsen a , Stefan Steuernagel c , Søren Balling Engelsen a a University of Copenhagen, Faculty of Life Sciences, Department of Food Science, Quality & Technology, Rolighedsvej 30,1958 Frederiksberg C, Denmark b Danisco A/S, Advanced Analysis, Edwin Rahrs Vej 38, 8220 Brabrand, Denmark c Bruker-Biospin GMBH, 76287 Rheinstetten, Germany article info Article history: Received 31 August 2008 Accepted 12 November 2008 Keywords: Alginate Calcium Nuclear magnetic resonance CP-MAS NMR HR-MAS NMR Chemometrics abstract The potential of using 1 H high-resolution (HR) magic angle spinning (MAS) nuclear magnetic resonance (NMR) of alginates suspended in D 2 O or 13 C cross-polarisation (CP) MAS NMR of alginate powders as an alternative method to the traditional 1 H solution-state NMR method for the analysis of the alginate monomer composition (mannuronate (M)/guluronate (G) ratio) has been investigated. The MAS NMR experiments can be performed directly on the intact alginate powders and thereby avoiding the rela- tively time-consuming and labour intensive sample preparation required for the traditional solution- state experiment. A total of 42 different sodium alginate samples were analysed and it was found that the M/G ratio derived from the HR-MAS NMR spectra were comparable to the M/G ratio calculated from the solution-state NMR spectra. However, both these methods overestimated the M/G ratio of samples with high calcium content. In contrast, the 13 C CP-MAS NMR spectra were not affected by residual calcium in the samples and thus provided more consistent measurements independent on physico-chemical phenomena such as partial solubility, viscosity and interactions with ions. It is concluded that 13 C CP- MAS NMR is a most suitable candidate for accurate analysis of the M/G ratio of intact alginate powders. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction Alginates are commercially extracted from brown seaweed (Phaeophyceae) and utilised in the food and pharmaceutical industries because of their gelling, viscosifying and stabilising properties (Draget, Moe, Skjåk-Bræk, & Smidsrød, 2006). From a structural point of view, alginates are linear polysaccharides composed of (1–4) linked b-D-mannuronic acid (M) and a-L- guluronic acid (G) arranged in a blockwise pattern along the chain as homopolymeric (MM or GG) or heteropolymeric (MG) regions. The M/G ratio and blockwise pattern vary according to season, age of population, species and geographic location (Haug, Larsen, & Smidsrød, 1974; Indergaard, Skjåk-Bræk, & Jensen, 1990; Stockton, Evans, Morris, Powell, & Rees, 1980) and are correlated to the functional properties of the alginates, i.e. solubility, interaction with metal ions, gel properties and viscosity (Haug, Myklesta, Larsen, & Smidsrød, 1967; Steginsky, Beale, Floss, & Mayer, 1992; Stokke, Smidsrød, Bruheim, & Skjåk-Bræk, 1991). Therefore, access to reliable methods for compositional analysis of alginates is essential, and for screening of large sample sets and quality control, rapid analytical methods requiring limited sample preparation are of great importance. 1 H nuclear magnetic resonance (NMR) spectroscopy has proven to be highly effective in the characterisation of alginates in solution (Grasdalen, 1983; Grasdalen, Larsen, & Smidsrød, 1979; Penman & Sanderson, 1972). However, at the concentrations suitable for an acceptable signal-to-noise ratio, alginate solutions are too viscous to give well-resolved spectra. Thus, the viscosity must be lowered by partial acid hydrolysis of the alginate chain, which is a relatively time-consuming and labour intensive procedure and perhaps more important the hydrolysis will alter the sample significantly compared to the intact sample. Another issue related to the analysis of alginates in solution is the possible selective micro-aggregation when divalent cations (e.g. calcium ions) are present. Calcium ions preferentially bind to the G-blocks (Draget et al., 2000; Grant, Morris, Rees, Smith, & Thom, 1973), resulting in G-block association and the formation of a gel network. This can give rise to a broad- ening of the 1 H NMR resonances originating from the G-units beyond the level of detection (Grasdalen, Larsen, & Smidsrød, 1981), resulting in an overestimation of the M/G ratio. * Corresponding author. University of Copenhagen, Faculty of Life Sciences, Department of Food Science, Quality & Technology, Rolighedsvej 30,1958 Freder- iksberg C, Denmark. Tel.: þ45 8943 5285, þ45 3533 3510; fax: þ45 8925 1077. E-mail addresses: tina.salomonsen@danisco.com, tisa@life.ku.dk (T. Salomonsen). Contents lists available at ScienceDirect Food Hydrocolloids journal homepage: www.elsevier.com/locate/foodhyd 0268-005X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodhyd.2008.11.009 Food Hydrocolloids 23 (2009) 1579–1586