Water Absorption of Freeze-Dried Meat at Different Water Activities: a Multianalytical Approach Using Sorption Isotherm, Differential Scanning Calorimetry, and Nuclear Magnetic Resonance LUCA VENTURI,PIETRO ROCCULI,CLAUDIO CAVANI,GIUSEPPE PLACUCCI, MARCO DALLA ROSA,* AND MAURO A. CREMONINI* Department of Food Science, University of Bologna, Campus of Food Science, P.zza Goidanich 60, 47023 Cesena, Italy Hydration of freeze-dried chicken breast meat was followed in the water activity range of a w ) 0.12 – 0.99 by a multianalytical approach comprising of sorption isotherm, differential scanning calorimetry (DSC), and nuclear magnetic resonance (NMR). The amount of frozen water and the shape of the T 2 -relaxogram were evaluated at each water content by DSC and NMR, respectively. Data revealed an agreement between sorption isotherm and DSC experiments about the onset of bulk water (a w ) 0.83–0.86), and NMR detected mobile water starting at a w ) 0.75. The origin of the short-transverse relaxation time part of the meat NMR signal was also reinvestigated through deuteration experiments and proposed to arise from protons belonging to plasticized matrix structures. It is proved both by D 2 O experiments and by gravimetry that the extra protons not contributing to the water content in the NMR experiments are about 6.4% of the total proton NMR CPMG signal of meat. KEYWORDS: NMR, DSC, sorption isotherm, freeze-dried meat, T 2 -relaxograms INTRODUCTION The mobility and availability of water in food systems depend on the extent of interactions between the aqueous phase and the biopolymers matrix (1). These parameters are of the utmost importance in food technology because the amount and physico- chemical behavior of water embedded in foods may trigger microbiological growth or even unwanted chemical reactions, thus lowering food quality and shelf life (2). It is thus highly desirable to attain a deep understanding of the interactions between water and food components to be able to produce clear- cut models and simple quality parameters that can be readily applied in the food industry. A partial solution to the problem of assessing the degree of availability of water in food materials has been known since the 1950s, when Scott and Salwin independently introduced the now well-known concept of “water activity” (a w ), whereby “boundness” to a food matrix is related to the relative vapor pressure of water (for a recent historical review see ref 3 and references therein); the studies on a w led to the description of a “food stability map” (4) that is still widely used by the food industry as a stability indicator for food quality control and shelf life prediction. Although it is common to refer to the mobility and availability of water in foods or hygroscopic polymers with the expression “state of water” (see, for example, refs 5–8) it must be borne in mind that, here, water is always as liquid as in the common liquid state, and it is held back by the capillary forces generated by the physical structure of the matrix beyond condensation. As simple as it is (a single parameter describes the status of the whole embedded water), a w suffers from a number of drawbacks that have been discussed in the literature along the years, many of them thoroughly reviewed in a famous paper by Slade and Levine (9). These researchers based their criticisms on the following points: (i) for a w to be a meaningful descriptor of the water status, it is necessary that at thermal equilibrium the partial vapor pressure above the food system is the same as that of the embedded water (i.e., thermodynamic equilibrium is reached). This condition is generally fulfilled in diluted food systems but is hardly met in concentrated food systems, owing to the low diffusion rate of water with respect to the time scale of measurement. In these systems only a kinetic steady state is reached, which is at the basis of the known hysteresis effect in sorption and desorption isotherms; (ii) even if thermodynamic equilibrium were reached, no way would exist for extracting meaningful information from the sorption or desorption iso- therms because the widely used BET (10) or GAB (11) equations are based on assumptions that do not hold good for food materials; (iii) a w is not an absolute food stability predictor because spoilage at a certain measured a w depends on food composition, physical structure, temperature, prior sample history, and even isotherm measurement methodology; (iv) a w defined as relative vapor pressure can reflect only the surface properties of a system but not necessarily the molecular * To whom correspondence should be addressed. E-mail: mauro.cremonini@unibo.it, marco.dallarosa@unibo.it. 10572 J. Agric. Food Chem. 2007, 55, 10572–10578 10.1021/jf072874b CCC: $37.00  2007 American Chemical Society Published on Web 11/30/2007