Color Degradation of Beans During Storage

Kinetics of color changes of beans (Phaseolus vulgaris) were investigated using sensory evaluation and objective colorimetric measurements during storage at elevated temperatures (25°–45°C). Whiteness parameter appeared to be accurately correlated to sensory evaluation data. It was found that browning depended strongly on the storage temperature and on the initial moisture content. An Arrhenius relationship was used to describe the effect of the temperature on the loss of quality. Both parameters indicated that the shelf life of beans may be extended by decreasing the initial moisture content of beans and by controlling storage conditions.

Keywords:

• Shelf life
• Legumes
• Accelerated storage stability test
• Color
• Browning reaction

INTRODUCTION

Loss of food quality during storage is one of the major problems for food manufacturers. Environmental conditions during storage usually vary significantly in warehouses without environmental control. The changes of environmental conditions constitute a difficulty in predicting the actual shelf life of a product. Legumes often become unacceptable due to significant quality loss and in some cases even mold growth. In these cases, the products were stored in an environment of high relative humidity (over 65%) and relatively high temperatures (above 25°C). Examination of quality loss during storage is certain to decrease the number of products return, resulting in a significant loss for food manufacturers.

A number of studies are reported in the literature[1] for the shelf life studies of various agricultural products such as peppers, onions, tomatoes, etc. Legumes and especially beans are commodities of high importance, since much of the world population staple relies on pulses as protein foods particularly in combination with cereals.[2] Different types of beans are consumed around the world. For example, in the Eastern Mediterranean, white dried beans are categorized according to size in medium, giant, and elephant. Despite the importance of beans, there are few shelf life data reported in the literature. Effort was devoted over the last several decades towards understanding the effect of storage on the quality of the cooked beans. It has been reported[3,4] that storage of dried beans resulted in deterioration of their quality, namely the water absorption capacity, texture of cooked beans, color, and flavor. Moreover, the storage of dried beans at elevated temperatures and humidity was found to reduce protein quality.[5]

Sensory properties of the legumes, such as, aroma, flavor, texture, and color are significant. Color of the packaged product, though, is the first quality parameter that the consumer perceives. Indeed, according to preliminary studies, color is the most important parameter describing the quality of beans. This is in agreement with various large-scale consumer studies in which 40–60% of grade points, related to quality, were assigned to color.[6,7]

The loss of quality during storage does not only affect the characteristics of the final product, but also the precooked product. The color of beans (especially the beans having white color) is changing during storage, affecting the desirability of the product by the consumers. More specifically, browning reactions that occur during storage could result in an unacceptable product. Colorimetric techniques have been used[8] to determine the reaction kinetics constants for the discoloration of vegetables. In a review work,[9] a number of research papers, which deal with quality control procedures using colorimeters for a number of different products, was reported. Color deterioration of foods, occurring during storage is due to either an enzymatic action or a chemical reaction. The enzymatic browning usually occurs in cases where enzymes are activated, often after a cut of the product, and results in very rapid change of the color. Non-enzymatic browning is due to either oxidative browning or non-oxidative browning in the absence of oxygen, as a result of chemical reactions between proteins and reducing sugars (Maillard reaction). Considering the typical shelf life of beans (1–2 years), the mechanism of color deterioration during storage is probably due to both enzymatic and chemical browning. The reaction rate of browning is due not only to composition (concentration and availability of protein, reducing sugars etc) but also to environmental factors (temperature and humidity).

Over the last decades, a number of studies have been reported that correlate non-enzymatic browning vs. the water activity and the temperature of various foods.[10] The importance of the state of starch-based food material in non-enzymatic browning has been demonstrated.[11] Non-enzymatic browning requires the existence of an amino group, and glucose that results in a macromolecule of characteristic brown color and flavor. The probability for this reaction to take place depends on the mobility of the whole molecules, or parts of the molecules. In the glassy state, the molecules are “frozen” and there is no translational or rotational motion of the molecules, only vibrational motion can exist.[11,12] Thus, the probability of a reaction taking place in the glassy state is very small when compared to the rubber state. The importance of water activity in the structural properties of foods has been investigated by a number of research groups.[11,13] Non-enzymatic browning is a reaction that leads to the deterioration of the food—not only due to the appearance of a characteristic brown color, but also due to the hardening of the product. Thus, the study of this phenomenon is very important for the pulses, especially for the white ones where the change of color is very evident. As was previously mentioned, the fundamental understanding of the factors that lead to the browning of legumes will be of major importance for the pulse industry, since it would reduce the unacceptable and rejected products. In various published papers, accelerated storage stability tests are reported to determine the shelf life of legumes and, in particular, beans. According to this methodology,[14] kinetic parameters of a deterioration reaction are determined by experimentation at extreme environmental conditions (temperature, humidity). Then, using an interpolation or extrapolation technique, one may predict the kinetic constant at the specified temperature. In this article, the study of the overall color deterioration of various types of white beans stored under various environmental conditions is reported. The objective of this research was to examine the storage stability of various types of pulses under different environmental conditions—close to those encountered in a typical warehouse and suggest a method to reduce the quality losses during storage.

MATERIALS AND METHODS

Legumes

Three types of pulses, differing in size and shape, packaged in polypropylene pouches were supplied by the local industry (EY.GE. Pistiolas, S.A.). The beans in order of increasing size were “medium”, “giants,” and “elephants”. Although chemical composition of the various bean types was found to be similar, initial moisture varied from lot to lot. Typically medium size pulses had lower moisture content (0.09–0.10 water/kg dry solids), while elephants and giant beans had the highest (0.15–0.16 kg water/kg dry solids).

Color Measurement Using an Objective Method

The color of the products was measured using a portable colorimeter (HunterLab MiniScan XE Plus, Hunter Associated Laboratories, Reston, VA, USA). The reflectance of the sample was measured as a function of the wavelength in the area of 400–700 nm. The spectrum was translated in the color constants L (lightness) and chromaticity parameters a and b of the Hunter Lab scale that are defined as follows:

Two standard tiles (one white and one black) were used for calibration of color before each set of measurements. A number of other parameters could result from the main color spectrum. The term whiteness or whiteness index (WI) is related to an area or volume in the color space where objects are considered to be white. The degree of whiteness is measured by the degree of distance of the object from a “perfect” white, and it is given automatically from the colorimeter apparatus.

Measurement of Moisture Content

Moisture content was determined using a vacuum oven method (70°C for 24 h). A small portion of product (around 5 g), removed from the storage room at various time intervals, was ground and used for moisture content determination.

Sensory Evaluation Tests

Sensory evaluation tests were performed in order to correlate the parameters estimated during the color measurements using the colorimeter with the product quality. The members of the panel group consisted of experienced members of a local pulse processing company, selected after preliminary tests. The color, aroma, and overall appearance of the samples were graded from the panel group, using a 1–9 hedonic scale. Grade 1 indicated an unacceptable sample while 9 implied a perfect product. Products that scored less than 5 were considered unacceptable in terms of quality.

Storage Stability Tests for Beans

Color change and moisture content were measured in a number of products stored at different temperature conditions. Samples consisted of 500 g of product packed in polyethylene film stored in temperature incubators (Binder Labortechnic, Tuttlingen, Germany) maintained at 25°, 35° and 45°C. Although the samples were hermetically sealed, moisture was transferred through the packaging material from the samples to the environment. Part of the range of the examined temperatures (25°–35°C) represents conditions often encountered in typical warehouses during the summer. The color of the samples was measured using objective (colorimeter) and sensory evaluation methods.

RESULTS AND DISCUSSION

The results are presented for each of the types of white beans examined in this study.

“Giant” Beans

As mentioned in the Materials and Methods section, a number of parameters were estimated when measuring the color using an objective method (a, b, L, WI). Quality would be described from the parameter having the strongest correlation with the scores of the sensory evaluation, will be used as index of the product quality. In Fig. 1(a)–(d) the parameters L, a, b, and WI describing the color of the sample versus the score of the sensory evaluation is shown. It may be seen from Fig. 1(a)–(c) that there is not a strong correlation between L, a, and b and the sensory evaluation score. However, in Fig. 1(d), it appears that the relationship between the WI and sensory evaluation score is such that WI could be used as an index for the product quality. This was expected, since the desired color of the product is white and WI indicates the perception of whiteness of the product. Similar observations were made for all types of white beans.

Figure 1. Correlation of the sensory evaluation score with color parameters in giant beans.

In Fig. 2, the WI of “giant” beans vs. storage time at temperatures of 25°C, 35°C, and 45°C is shown. Storage of samples at 25°C and 35°C did not result in a significant color change when compared to the change occurring at 45°C. As it was previously mentioned, the rate of browning reaction depends on the state of the sample and particularly on the mobility of the reactants.[11] The small change in WI value at 25°C and 35°C, indicates that the product was in the glassy state. Thus, the mobility of the molecules was very limited, and no color modification was observed. If one considers that browning is a zero order reaction, at 45°C the value of reaction rate constant (0.302 days⁻¹) indicated that the sample was not in the glassy state. At 45°C due to the higher availability of water (higher water activity for the same moisture content), the samples were probably at a temperature higher than the glass transition temperature (Tg). Therefore the samples were not in the glassy state, and macromolecules, or parts of them can be mobile, and the browning reaction may take place.

Figure 2. Whiteness vs. storage time for giant beans at three temperatures (25°C, 35°C, 45°C).

It may be noted that after the 50^th day of storage, no significant change in the color of the samples was observed. Experiments were performed under realistic storage conditions, and moisture loss could have occurred. As it was stated above, the giant beans contained higher water content than the medium type due to inadequate drying in the field. In Fig. 3, the change of moisture content of giant beans during storage is shown. Similar results were obtained for all types of beans examined. From Fig. 3, it may be concluded that there is a significant moisture loss during storage. The glass transition temperature (Tg) is a strong function of the water activity.[15] Thus, when the water activity of the samples was reduced, i.e., when the moisture content of the sample was reduced, the Tg increased sharply. At the initial stage, the sample was not probably in the glassy state due to the relatively high moisture content in the sample; therefore the browning reaction was feasible. After approximately the 50^th day of storage, and due to moisture loss, a phase transition could have occurred resulting in a sharp decrease in the browning rate.

Figure 3. Drying of giant beans due to storage at 45°C.

Using the values of whiteness during storage it was possible to estimate the order and the constant of the reaction. Since the reactants of the browning reaction (amino-groups, glucose, oxygen) were in excess, no change in the concentration of the reactants was expected.[16] A zero-th order rate of reaction may take place during non-enzymatic browning, although more complex kinetic models may apply for some sugar-amino groups reactions. Therefore one should expect that the overall reaction kinetics for browning is zero-th order, more specifically:

(1)

where, WI is the whiteness index, WI_o is the initial value of the whiteness index, t is the storage time and k is the constant of the reaction. The reaction rate constant (k) is usually a function of temperature and water activity. Thus, one should ultimately keep the moisture content of the samples constant throughout the experiments. The experiments, though, were designed keeping in mind storage conditions in a typical warehouse, where partial drying of samples occurs. Therefore, the values of the whiteness measured after the 50^th day should be ignored in the calculation of k. In Fig. 4 the experimental and the values of k, resulting from the zero-order reaction, are presented. The values of k were calculated using linear regression. The value of R² is 0.99, indicating that the order of the reaction was indeed of zero order.

Figure 4. Kinetics rate of whiteness change during storage of “giant” beans at 45 °C.

“Elephant” Beans

Experiments similar to those in giant beans were performed using the larger “elephant” beans. The chemical composition of the “elephant” and “giant” beans was similar, the moisture content of the two samples though was quite different (elephants contained 3–5% higher moisture content than giants). Similarly to the “giant” beans, the parameter used as an indicator of the color was found to be the whiteness.

In Fig. 5, the whiteness index (WI) vs. storage time at temperatures of 25°C, 35°C, and 45°C as in the case of “giant” beans is shown for elephant beans. One can see that WI was changed not only during storage at 45°C but also at 35°C and 25°C. This difference with the observations made for “giant” beans was due to the difference of the moisture content of the samples. While “giant” beans were probably in the glassy state at 35°C and 25°C the “elephant” beans that had higher moisture content were possibly in the rubbery state for all employed temperatures. Therefore, it is expected that the color of “elephant” beans would change even at 25°C due to the higher mobility of the reactants, similar results have been reported in the literature.[17] The changes in the moisture content and non-enzymatic browning have been correlated well with the relative humidity of the place of storage for texturized soya protein products.

Figure 5. Whiteness vs. storage time for elephant beans for three storage temperatures (25°C, 35°C, 45°C).

Similarly to the “giant” beans, a zero-th reaction rate was used to describe the phenomenon of color deterioration. The values of WI measured after the 50^th day were omitted for reasons that were previously explained. The values of the reaction rate are presented in Table 1. As it was expected, the color deterioration was faster at higher temperatures. The temperature dependence of the reaction constant is given by of an Arrhenius type of relationship and is as follows:

(2)

Table 1 Calculated reaction constants values (k) at various storage temperatures (25°C, 35°C, and 45°C) for elephant beans

T (K)	k (day^−1)
298	0.08
308	0.40
318	1.00

Where, A is the frequency factor, Ea is the activation energy, R is the universal gas constant, and T is the absolute storage temperature. An Arrhenius plot is shown in Fig. 6. Using a linear regression, the parameters A and Ea were found to be equal to A = 2.7 10¹⁶ day⁻¹ and Ea = 99.8 kJ/mol. The relatively high value of the activation energy implies that the browning reaction is indeed a chemical reaction. Activation energies of the same order were reported[18,19] for browning reactions.

Figure 6. Arrhenius plot showing the effect of temperature on the kinetic rate constant of browning reaction of stored “elephant” beans.

Medium Size Beans

The effect of storage time in the color of medium size beans was also studied. In Fig. 7 the WI of medium size beans vs. storage time is shown at temperatures of 25°, 35°, and 45°C for medium size beans. Color alteration was observed for all temperatures. The reaction constants for the three different temperatures were also estimated and are shown in Table 2. The activation energy, Ea, was found to be equal to 121.9 kJ/mol, of the same order of magnitude, but higher than the one calculated for the elephant beans.

Figure 7. Whiteness vs. storage time for “medium” white beans stored at three temperatures.

Table 2 Calculated reaction constants values (k) at various storage temperatures (25°C, 35°C, and 45°C) for medium size beans

T (K)	k (day^−1)
298	0.014
308	0.103
318	0.302

In Fig. 8, the whiteness vs. storage time for three different types of beans stored at 45°C is shown. While “elephant” and “giant” beans had a similar behavior, there is a big difference with the medium size beans. This was not surprising since the moisture content of medium size beans (9–10% on dry basis) is much lower than the moisture content of the rest of the beans. It seems that the moisture content of beans is very important, since high moisture content may result in excessive browning. It seems from the above experiments that an initial moisture content of around 0.11 kg water/kg dry solids is critical because under this value the quality is attained for an extended storage time, even when the storage temperature is relatively high (over 35°C).

Figure 8. Whiteness vs. storage time for storage for three different types of white beans stored at 45°C.

CONCLUSIONS

The browning of beans depended strongly on the storage temperature and on the initial moisture content of stored dried beans. Both parameters indicate that the shelf life of beans may be extended by decreasing the initial moisture content of beans and by controlling the temperature and relative humidity of the storage facility. A zero-th order of reaction was found to fit the browning reaction for beans. The browning depended strongly on the storage temperature and also on the initial moisture content of stored beans. An Arrhenius relationship was used to describe the effect of the temperature. Initial moisture content of appeared to be critical. Although the exact value of the critical moisture content has yet to be determined—through experiments in controlled relative humidity environment—it appears that products with initial moisture content less than 0.11 kg water/kg dry solids have a long shelf life.

ACKNOWLEDGMENT

Part of this research was supported by the Greek Ministry of Development (General Secretariat of Research and Technology), Program #97BE212.