Developments in glass transition determination in foods using moisture sorption isotherms Brady P. Carter a,⇑ , Shelly J. Schmidt b,1 a Decagon Devices, Inc., 2365 Northeast Hopkins Court, Pullman, WA 99163, United States b Department of Food Science and Human Nutrition, University of Illinois at Urbana-Champaign, 367 Bevier Hall, 905 South Goodwin Ave., Urbana, IL 61820, United States article info Article history: Available online 22 June 2011 Keywords: Glass transition Glass transition temperature Moisture sorption isotherms Dynamic isotherms Critical water activity abstract The food polymer science (FPS) approach has been effectively used to investigate the physical stability of amorphous food materials. The glass transition, a key FPS parameter, has traditionally been deter- mined using thermal techniques that scan temperature while holding the plasticizer (moisture) content constant. Moisture sorption isotherms provide information about the physical properties of food as the plasticizer level is adjusted and temperature is held constant. New automatic isotherm generators can be used to produce high resolution, dynamic isotherms much faster than traditional static methods. Dynamic isotherms for a small selection of amorphous materials have been investi- gated and shown to experience distinct inflection points in the water activity region where the glass transition temperature is close to the experimental temperature. Several studies on amorphous spray dried milk powder and amorphous polydextrose indicate very good agreement between glass transi- tions determined using thermal techniques and dynamic isotherm methods. This agreement suggests that dynamic isotherms are a viable alternative to traditional thermal methods for investigating glass transitions of amorphous foods. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction A glass transition can be described as a change that occurs in amorphous materials from a high viscosity, ‘‘frozen’’ glassy state to a lower viscosity, rubbery state (Roos, 2007). A material in the glassy state behaves like a brittle solid, but without crystalline structure and only short ranges of order. The glass transition con- cepts that have long been understood in the field of polymer sci- ence can also be applied to food polymers (Slade & Levine, 1987; Sperling, 1986). A phase transition from glassy to rubbery results in drastic changes in molecular mobility of food polymers, which has been linked to changes in product quality and can result in a loss of stability for low moisture amorphous foods (Roos, 2007; Roos & Karel, 1991a; Slade & Levine, 1991; White & Cakebread, 1966). Foods in the glassy amorphous state exist in a metastable condition and remain stable for extended periods of time (months to years); however, once they have transitioned into the rubbery state, all rates of time dependent quality loss processes increase and shelf life decreases to weeks, days, or even hours (Roos, 2007). Key physical stability factors identified to be linked to the glass transition are an increased susceptibility to stickiness, caking, collapse and crystallization (Aguilera, Levi, & Karel, 1993; Haque & Roos, 2004; Jouppila, Kansikas, & Roos, 1997; Jouppila & Roos, 1994; Paterson, Brooks, Bronlund, & Foster, 2005; Roos, 2002; Roos & Karel, 1992). The important parameter in glassy to rubbery phase transitions is the glass transition temperature (Tg). Substantial efforts have been made to link Tg, or more specifically, T–Tg, to key stability loss events and to model the effect using Williams, Landel, and Fer- ry (WLF) (Williams, Landel, & Ferry, 1955) or similar types of mod- els (Aguilera et al., 1993; Haque & Roos, 2004; Jouppila & Roos, 1994; Jouppila et al., 1997; Paterson et al., 2005; Roos, 2002; Roos & Karel, 1992). The glass transition can roughly be categorised as a second order phase change that is accompanied by thermodynamic changes in enthalpy, changes in dielectric properties, and mechan- ical changes. It is worth noting that the glass transition events measured using current methods are all kinetic events that depend on the time scale of the experimental method, which is in contrast with a true thermodynamic second order event (Schmidt, 2004). This would indicate that the methods measure some time depen- dent consequence of the glass transition and not necessarily the transition itself. Common methods for investigating glass transitions have fo- cused on identifying thermodynamic, mechanical, or dielectric changes while scanning temperature to identify the Tg. This testing 0308-8146/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodchem.2011.06.022 ⇑ Corresponding author. Tel.: +1 509 332 2756; fax: +1 509 332 5158. E-mail addresses: brady@decagon.com (B.P. Carter), sjs@illinois.edu (S.J. Schmidt). 1 Tel.: +1 217 333 6963. Food Chemistry 132 (2012) 1693–1698 Contents lists available at ScienceDirect Food Chemistry journal homepage: www.elsevier.com/locate/foodchem