Estimation of apparent glass transition temperature from the release of 1-methylcyclopropene included in α-cyclodextrin

Abstract

The glass transition temperature (T_g) is an important physical property affecting the release of 1-methylcyclopropene (1-MCP) from its inclusion complex with α-cyclodextrin (α-CD). The release of 1-MCP from α-CD was investigated using humidity ramping at constant temperature and temperature ramping at constant humidity. The apparent T_g values of α-CD were estimated from the collapse temperatures observed in the release experiments and compared with the temperatures calculated using the Gordon–Taylor equation. Good agreement between the estimated and calculated T_g values indicated that the ramping method for flavor release from a powder matrix would be a simple and effective approach to estimate the T_g of the matrix.

Introduction

α-Cyclodextrin (α-CD) is a cyclic oligosaccharide that forms inclusion complexes with small molecules accommodated in its hollow truncated cone structure. α-CD can encapsulate gaseous compounds such as carbon dioxide (Neoh et al., 2006; Ho et al., 2015), 1-methylcyclopropene (1-MCP) (Neoh et al., 2007; Ariyanto and Yoshii, 2019), and ethylene (Ho et al., 2011). 1-MCP is an ethylene antagonist that is effective for extending the shelf life of fruits (Blankenship and Dole, 2003; Watkins, 2006). Furthermore, powdery 1-MCP encapsulated in α-CD can effectively maintain the freshness of vegetables and fruits during the distribution process. The ability of 1-MCP to delay ripening, senescence, and decay of climacteric fruit by blocking the action of ethylene has been demonstrated in many studies (Liu et al., 2015; Barry and Giovannoni, 2007; Lelièvre, et al., 1997; Saltveit, 1999). Temperature and humidity are the two main factors that affect the release of 1-MCP from its inclusion complexes with α-CD powder. Neoh et al. (2008) showed that 1-MCP encapsulated with α-CD began to dissociate at 120 °C under dry conditions, and that the click point temperature, where the temperature dependence of the dissociation constant changed stepwise, was 156 °C by both differential scanning calorimetric and thermogravimetric analyses. At high relative humidity (RH) values, the fine particles of α-CD powder encapsulating 1-MCP aggregate and slow down the release of 1-MCP (Neoh et al., 2008; Neoh et al., 2017). The structural changes in the α-CD powder encapsulating 1-MCP are related to their collapse process. Ariyanto and Yoshii (2019) reported that the collapse of an α-CD powder encapsulating 1-MCP occurred at RH values higher than 60 % and reduced the release of 1-MCP from the powder. In addition, the decrease in the 1-MCP release rate was attributed to the glass transition and collapse of α-CD. Therefore, the collapse and glass transition of a flavor-including α-CD powder are important phenomena for estimating the release behavior of the flavor under humid conditions. Under humid conditions, a guest compound such as a flavor encapsulated in α-CD is replaced with a water molecule at a certain temperature to begin to be released. In this study, the temperature is defined as the apparent glass transition temperature.

Caking of amorphous food powders results from their phase transition from a glassy to a less viscous liquid-like state; this process was termed “humidity caking” by Peleg (1983). Powder caking significantly affects the flavor release rate from the powder, owing to powder aggregation. The process causes a sudden decrease in the flavor release rate, denoted as “collapse”. To and Flink (1978) found similarities between the collapse temperatures and T_g values of polymers. Carolina et al. (2007) investigated the collapse occurrence and loss of aroma strength in encapsulated strawberry and orange flavors after storage at various relative humidities, with the aim to determine the glass transition temperature of commercial spray-dried flavor powders. Righetto and Netto (2005) also investigated the glass transition and stability of juice from encapsulated immature acerola. Stickiness was observed for the encapsulated juices at temperatures close to T_g, and collapse occurred at temperatures of 20 °C or more above T_g. Roos and Karel (1991) observed the collapse of maltodextrins (MDs, with dextrose equivalent[DE] values of 10–25) at temperatures of 40–70 °C above T_g. Reineccius (1995) correlated the retention of flavor with the T_g and the collapse temperature of the wall materials. Gunning et al. (1999) also indicated that the release and oxidation of encapsulated flavors are related to the collapse and crystallization of the powder matrix. Water sorption plays a crucial role in the crystalline structure and stability of the encapsulated powder. Yamamoto et al. (2012) applied the dynamic vapor system (DVS) method to investigate the release of d-limonene encapsulated in spray-dried cyclodextrin powder. An extensive review of powder caking has been published by Zafar et al. (2017), who discussed the bulk powder caking phenomenon and reported its dependence on the material properties, environmental conditions, and process dynamics. They showed that amorphous materials are susceptible to caking because of the influence of the environmental conditions on the glass transition temperature.

We aim to develop storage packaging in which 1-MCP encapsulated in α-CD is continuously released, even if the temperature and humidity change, so that fruits and vegetables can be stored for a long period of time. As mentioned above, the apparent T_g of the 1-MCP/α-CD complex largely affects the release rate of 1-MCP. In this context, the apparent T_g was estimated from the change in 1-MCP release flux using humidity ramping at constant temperature and temperature ramping at constant humidity.

Materials and Methods

Materials.    Gaseous 1-MCP was produced according to a previously reported method (Ariyanto and Yoshii, 2019). The chemicals used for the synthesis of 1-MCP were purchased from FUJIFILM Wako Pure Chemical and Sigma-Aldrich Japan (Tokyo, Japan). Isobutylene standard gas (100 µL/L) was purchased from Sumitomo Seika Chemical (Osaka, Japan). α-CD (99 %) was obtained from Cyclochem (Kobe, Japan).

Release experiments.    Release experiments of 1-MCP from α-CD powder were carried out using a hand-built dynamic vapor sorption (DVS) system (Yamamoto et al., 2012); the RH was increased from 10 to 90 % at a constant temperature of 30, 40, or 50 °C and the temperature was ramped from 20 to 60 °C at a constant RH of 65, 70, or 80 %. The procedure and experimental DVS approach followed those of Yamamoto et al. (2012) and Neoh et al. (2017). Humidity ramping was realized by increasing the temperature of a water bath in which nitrogen was passed. The temperature of the vessel containing the powder was varied using a DSSP93 controller (Shimaden, Tokyo, Japan), in such a way that the humidity increased linearly from 10 to 90 % in 4 h. Approximately 30 mg of the α-CD powder encapsulating 1-MCP was placed into a flat-bottomed aluminum sample pan (1 mm × 13 mm i.d.). The sample holder was placed in a glass release vessel (78 mm × 15 mm i.d.), whose temperature was maintained at the desired value. In the temperature-ramping experiments at constant humidity, the temperature of the vessel containing the powder was increased linearly from 20 to 60 °C in 4 h using a water bath as follows: the temperature of powder vessels placed in an incubator (IJ300W, Yamato Scientific Co., Ltd., Tokyo, Japan) was set to 20 °C. The temperature of the water bath through which nitrogen flowed to regulate the powder temperature was set to 13.2, 14.5 or 16.5 °C, and then it rose linearly at a rate of 0.945, 0.948 or 0.970 °C/min. By this operation, the humidity in the powder vessel was controlled to be 64.8 to 65.1 %, 70.6 to 69.8 %, or a constant RH, 80.2 % during the heating process. The release of 1-MCP from the α-CD powder was monitored using a gas chromatograph (Shimadzu GC-14B; Shimadzu Corp., Kyoto, Japan) equipped with a column (2.1 m × 3.2 mm i.d.) packed with PEG-20M (20 % on Chromosorb W 80/100 AW mesh; Shinwa Chemical Industries, Kyoto, Japan) and a flame ionization detector. The injection and detector temperatures were set at 140 and 200 °C, respectively, whereas the column temperature was maintained at 130 °C. The humidity in the nitrogen stream was monitored using a HMP233 sensor (Vaisala, Helsinki, Finland). For each condition, the release experiment was performed in duplicate and the results were averaged.

Results and Discussion

Fig. 1 shows the flux of 1-MCP released from the α-CD powder during humidity ramping from 10 to 90 % RH at 30, 40, and 50 °C. The flux was defined as the ratio of the amount of 1-MCP released per unit time to the total 1-MCP amount included in the α-CD powder. The arrows in the figure indicate the points at which the flux began to decrease. The flux increased gradually starting from 40 % RH at 30 °C, 30 % RH at 40 °C, and 20 % RH at 50 °C. The RH values at the first point where the flux began to decrease were 65.7 % at 30 °C, 47.3 % at 40 °C, and 42.5 % at 50 °C; the powder might have started to aggregate at these points. The RH values at the second point of decrease in the flux, which reflect the humidity conditions at which collapse occurred, were 81.7 % at 30 °C, 76.9 % RH at 40 °C, and 62.5 % RH at 50 °C. The release rates of 1-MCP decreased at the humidity conditions. These change points were called the collapse of the inclusion complex. Powder aggregation caused collapse, resulting in coating of the powder surface. Neoh et al. (2008) also observed the collapse of α-CD powder encapsulating 1-MCP by SEM. At all temperatures, the 1-MCP release rates showed another sharp increase at 85 % RH, as the α-CD powder encapsulating 1-MCP became moist and fluid.

Fig. 1.

1-MCP flux released from α-CD powder during humidity ramping from 10 % to 90 % RH at (□) 30 °C, (△) 40 °C, and (○) 50 °C. Solid and dotted arrows indicate the points where collapse of the powder matrix occurred. The dotted arrow and solid arrow indicated the aggregation of the inclusion complex powder begun and the collapse occurred.

Fig. 2 shows the flux of 1-MCP released from the α-CD powder during temperature ramping at 0.17 °C/min from 20 to 60 °C at a constant RH of 65, 70, or 80 %. The 1-MCP flux released at 65 % RH started to decrease slowly at 20 °C and increased again at 27 °C. The flux at a constant humidity of 65 % RH reached its maximum value at 36 °C and then decreased with increasing temperature. At 70 % RH, the flux started to increase from 20 °C, reaching its maximum value at 26.8 °C. At higher temperatures, the flux gradually decreased and reached a lower value of approximately 1.4 × 10⁻⁶ s⁻¹ at 35 °C. The shape of the decreasing curve at 70 % RH was similar to that at 65 % RH. Moreover, the 1-MCP flux rapidly decreased to almost zero between 35 and 60 °C. At 80 % RH, the 1-MCP flux decreased steeply from 20 to 22 °C and then gradually increased to its maximum value at 41 °C. Subsequently, the flux decreased from 17.4 × 10⁻⁶ to approximately 5 × 10⁻⁶ s⁻¹ at 42 °C and then leveled off between 42 and 60 °C. In the temperature ramping experiments, the maximum fluxes of 1-MCP released at 65, 70, and 80 % RH were almost the same, with values around 17 × 10⁻⁶ s⁻¹. These values were lower than those observed in the humidity ramping experiments shown in Fig. 1. The lower fluxes might be due to the higher initial moisture contents in the temperature ramping experiment at 65, 70, and 80 % RH compared to those of the humidity ramping tests at 30, 40, and 50 °C. Shrestha et al. (2007) used the temperature ramping method to investigate the physical properties of spray-dried powders with MDs of various DEs using thermal mechanical compression tests. Sultana et al. (2019) also adopted the simple method to estimate the apparent T_g of spray-dried powders based on their flavor-release behavior. Although many studies have used the temperature ramping method to estimate the physical properties of food powders, there have been no investigations on the flavor release from powders based on temperature ramping at constant humidity, since this approach requires control of both the humidity and temperature of the incubation vessel.

Fig. 2.

1-MCP flux released from α-CD powder during temperature ramping at 0.17 °C/min from 20 to 60 °C at constant RH values of (□) 65 %, (△) 70 %, and (○) 80 %. The arrows have the same purpose as those in Fig. 1.

Fig. 3 shows the apparent T_g estimated from the collapse points in the humidity and temperature ramping experiments (Figs. 1 and 2). The circles represent the T_g values estimated from the temperature at which the collapse occurred at 65, 70, and 80 % RH. The open and closed squares represent the first and second points where collapse occurred in the humidity ramping experiments at 30, 40, and 50 °C, respectively. The dotted square marks the small collapse point observed in the humidity ramping experiment at 40 °C. The solid and dotted curves were calculated using the Gordon-Taylor equation, based on the moisture sorption isotherms at 50 and 30 °C, respectively (Hunter et al., 2010):   

where T_a and T_b are the glass transition temperatures of α-CD and water, respectively; T_a was assumed to be 150 °C since the onset transition temperature of the dissociation reaction of 1-MCP/α-CD was 150 °C (Neoh et al., 2008). T_b was set to −135 °C (Roos, 1995); ω_a and ω_b are the weight fraction of α-CD and the moisture content, respectively, whereas k is a constant, which was set to 6. The solid squares at 40 and 50 °C are located above the curves calculated using the Gordon–Taylor equation, whereas the other symbols are scattered above and below the curves. These results indicate that the temperature and humidity at which collapse occurs in humidity or temperature ramping experiments of flavor release from a powder such as α-CD can be used to estimate the apparent T_g.

Fig. 3.

Apparent glass transition temperatures estimated (○) in temperature ramping experiments at 65 %, 70 %, and 80 % RH and (□, ■) in humidity ramping experiments at 30, 40, and 50 °C. The open and filled squares represent the T_g values estimated at the lower and higher humidity conditions. The dotted square represents the T_g estimated from the small collapse in the humidity ramping experiment at 40 °C. The solid and dotted curves were calculated by the Gordon–Taylor equation using the moisture sorption isotherms of α-CD at 50 and 30 °C, respectively (Hunter et al., 2010).

Peleg (1983) described these caking phenomena as humidity caking. Gunning et al. (1999) reported the factors affecting the release of flavors encapsulated in carbohydrate matrices; they indicated that increased releases of acetaldehyde and benzaldehyde/benzyl alcohol from a sucrose-maltodextrin (DE = 10) matrix could be observed at approximately T_g and T−T_g = 10 °C (where T is the storage temperature), respectively. Soottitantawat et al. (2004) also presented the release rate constants as a function of T−T_g for the release of d-limonene from different matrices, such as mixtures of MD with gum arabic, HiCap 100^® modified starch, and soybean soluble polysaccharides, as well as HiCap 100^® alone. They reported that the release rate constant decreased near T_g because the collapse of the powders around this temperature resulted in the particles aggregating and adhering to each other, which in turn led to closing of the spaces between the particles and a decreasing surface area for d-limonene release. Carolina et al. (2007) reported glass transition data to correlate the collapse occurrence and the reduction in aroma intensity for encapsulated strawberry and orange flavors stored at various RHs. Li and Schmidt (2011) applied ramping and equilibrium water vapor sorption methods to determine the critical relative humidity at which the glassy to rubbery transition occurred in polydextrose. They showed that the water sorption method with humidity ramping could be used to predict the conditions corresponding to the occurrence of the glassy to rubbery transition. These reports suggested that the flavor release behavior changed near T_g.

Sultana et al. (2019) estimated the apparent T_g of spray-dried and yeast powders based on the flavor release behavior under temperature ramping conditions. They suggested that the flavor release rate from the powders depended on the T_g and the collapse point at which a sudden change in flavor release rate occurred, which allowed estimation of the T_g of the powder encapsulating the flavor. We have previously reported on paper coating (Ariyanto and Yoshii, 2019) or polymer coating (Yamamoto et al., 2012) of α-CD powders encapsulating 1-MCP for the development of packaging to extend the shelf life of fruits and vegetables. The results obtained in the present study are useful for developing such packaging, since the factors that influence the release behavior of 1-MCP were clarified.

Acknowledgements    Part of this study was supported by JSPS KAKENHI (Grant Number 21K05464) and the Japan Food Chemical Research Foundation.

Conflict of interest    There are no conflicts of interest to declare.

References

•  Ariyanto,  H. D. and  Yoshii,  H. (2019). Effect of stepwise humidity change on the release rate constant of 1-methylcyclopropene (1-MCP) in a cyclodextrin inclusion complex powder. Food Packag. Shelf Life, 21, 100322.
•  Barry,  C. S. and  Giovannoni,  J. J. (2007). Ethylene and fruit ripening ethylene and fruit ripening. J Plant Growth Regul., 26, 143-159.
•  Blankenship,  S. M. and  Dole,  J. M. (2003). 1-Methylcyclopropene: a review. Postharvest Biol Technol., 28, 1-25.
•  Carolina,  B. C.,  Carolina,  S.,  Zamora,  M. C., and  Jorge,  C. (2007). Glass transition temperatures and some physical and sensory changes in stored spray-dried encapsulated flavors. LWT - Food Sci. Technol., 40, 1792-1797.
•  Gunning,  Y. M.,  Gunning,  P. A.,  Kemsley,  E. K.,  Parker,  R.,  Ring,  S. G.,  Wilson,  R. H., and  Blake,  A.. (1999). Factors affecting the release of flavor encapsulated in carbohydrate matrixes. J. Agric. Food Chem., 47, 5198-5205.
•  Ho,  B. T.,  Joyce,  D. C., and  Bhandari,  B. R. (2011). Encapsulation of ethylene gas into α-cyclodextrin and characterisation of the inclusion complexes. Food Chem., 127, 572-580.
•  Ho,  T. M.,  Howes,  T., and  Bhandari,  B. R. (2015). Encapsulation of CO₂ into amorphous and crystalline α-cyclodextrin powders and the characterization of the complexes formed. Food Chem., 187, 407-415.
•  Hunter,  N. E.,  Frampton,  C. S.,  Craig,  D. Q. M., and  Belton,  P. S. (2010). The use of dynamic vapour sorption methods for the characterisation of water uptake in amorphous trehalose. Carbohydr. Res., 345, 1938-1944.
•  Lelièvre,  J. M.,  Latchè,  A.,  Jones,  B.,  Bouzayen,  M., and  Pech,  J., C. (1997). Ethylene and fruit ripening. Physiol. Plant., 101, 727-739.
•  Li,  Q. E., and  Schmidt,  S. J. (2011). Use of ramping and equilibrium water vapor sorption methods to determine the critical relative humidity at which the glassy to rubbery transition occurs in polydextrose. J. Food Sci., 76, 149-157.
•  Liu,  M.  Pirrello,  J.,  Chervin,  C.,  Roustan,  J. P., and  Bouzayen,  M. (2015). Ethylene control of fruit ripening: revisiting the complex network of transcriptional regulation. Plant Physiol., 169, 2380-2390.
•  Neoh,  T. L.,  Ariyanto,  H. D.,  Menéndez Galvan,  P., and  Yoshii,  H. (2017). Controlled release of 1-methylcyclopropene from its functionalised electrospun fibres under constant and linearly ramped humidity. Food Addit. Contam., Part A, 34, 1690-1702.
•  Neoh,  T. L.,  Yamauchi,  K.,  Yoshii,  H., and  Furuta,  T. (2007). Kinetics of molecular encapsulation of 1-methylcyclopropene into alpha cyclodextrin. J. Agric. Food Chem., 55, 11020-11026.
•  Neoh,  T. L.,  Yamauchi,  K.,  Yoshii,  H., and  Furuta,  T. (2008). Kinetic study of thermally stimulated dissociation of inclusion complex of 1-methylcyclopropene with α-cyclodextrin by thermal analysis. J. Phys. Chem., 112, 15914-15920.
•  Neoh,  T. L.,  Yoshii,  H., and  Furuta,  T. (2006). Encapsulation and release characteristics of carbon dioxide in α-cyclodextrin. J. Incl. Phenom. Macrocycl. Chemis., 56, 125-133.
•  Peleg,  M. (1983). Physical characteristics of food powders. In “Physical Properties of Foods,” ed. by  M.  Peleg, and  E. B.  Bagley. AVI, Westport, pp. 293-321.
•  Reineccius,  G. A. (1995). Controlled-release techniques in the food-industry. In “Encapsulation and Controlled Release of Food Ingredients,” ACS symposium series no. 590, ed. By  S. J.  Risch and  G. A.  Reineccius. ACS, Washington, pp. 8-25.
•  Righetto,  A. M. and  Netto,  F. M. (2005). Effect of encapsulating materials on water sorption, glass transition and stability of juice from immature acerola. Int. J. Food Prop., 8, 337-346.
•  Roos,  Y. (1995) Phase transitions in foods. Academic press. San Diego, CA. p. 159.
•  Roos,  Y. and  Karel,  M. (1991). Phase transitions of mixtures of amorphous polysaccharides and sugars. Biotechnol. Prog., 7, 49-53.
•  Saltveit,  M. E. (1999). Effect of ethylene on quality of fresh fruits and vegetables. Postharvest Biol. Technol., 15, 279-292.
•  Shrestha,  A. K.,  Ua-Arak,  T.,  Adhikari,  B. P.,  Howes,  T., and  Bhandari,  B. R. (2007). Glass transition behavior of spray dried orange juice powder measured by differential scanning calorimetry (DSC) and thermal mechanical compression test (TMCT). Int. J. Food Prop., 10, 661-673.
•  Soottitantawat,  A.,  Yoshii,  H.,  Furuta,  T.,  Ohgawara,  M.,  Forssell,  P.,  Partanen,  R.,  Poutanen,  K., and  Linko,  P. (2004). Effect of water activity on the release characteristics and oxidative stability of d-limonene encapsulated by spray drying. J. Agric. Food Chem., 52, 1269-1276.
•  Sultana,  A.,  Zhu,  J. L., and  Yoshii,  H. (2019). Estimation of the apparent glass transition temperature of spray-dried emulsified powders and yeast powders by their flavor release behavior under temperature ramping conditions. Presented at 21th International Drying Symposium, Valensia, September 11–14, pp. 803-809.
•  To,  E. C. and  Flink,  J. M. (1978). ‘Collapse’, a structural transition in freeze dried carbohydrates: II. Effect of solute composition. Int. J. Food Sci. Technol., 13, 567-581.
•  Watkins,  C. B. (2006). The use of 1-methylcyclopropene (1-MCP) on fruits and vegetables. Biotechnol. Adv., 24, 389-409.
•  Yamamoto,  C.,  Neoh,  T. L.,  Honbou,  Y.,  Furuta,  T.,  Kimura,  S., and  Yoshii,  H. (2012). Evaluation of flavor release from spray-dried powder by ramping with dynamic vapor sorption-gas chromatography. Drying Technol., 30, 1045-1050.
•  Yamamoto,  C.,  Neoh,  T. L.,  Honbou,  H.,  Yoshii,  H., and  Furuta,  T. (2012). Kinetic analysis and evaluation of controlled release of D-limonene encapsulated in spray-dried cyclodextrin powder under linearly ramped humidity. Drying Technol., 30, 1283-1291.
•  Zafar,  U.,  Vivacqua,  V.,  Calvert,  G.,  Ghadiri,  M., and  Cleaver,  J. A. S. (2017). A review of bulk powder caking. Powder Technol., 313, 389-401.