Sorption isotherms of ingredients and diets for poultry

This study aimed to determine sorption isotherms of ingredient and poultry diet. The samples were encapsulated in capsules and dehydrated by oven-drying in a desiccator for more than 24 Research, Society and Development, v. 9, n. 9, e828997729, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7729 2 hours. The samples were transferred to desiccator containing water in the base and placed in the oven, with one sample of each material being removed at incremental intervals. The sample was weighed and for determination of water activity and for dry matter. The moisture and water activity data were evaluated by eight mathematical models. The GAB mathematical model fitted the experimental data to constitute the isotherm for each material. Type II sorption isotherms were found, except for BHT: demonstrated values that did not fit the isotherm determination. The hygroscopic behavior of the ingredients in ascending order were: Lthreonine, limestone, BHT, DLmethionine, L-valine, Ltryptophan, phosphate, kaolin, vitamin supplement, salt, mycotoxin deactivator, pelleted rooster diet, mash rooster diet, mash layer diet, pelleted layer diet, corn, bacitracin zinc, vitamin mineral supplement, phytase, rice bran, wheat bran, mineral supplement, soybean meal, coccidiostat, LLysine HCl and choline chloride. Ingredients and diets have different hygroscopic behavior: can lead to deterioration and low accuracy in nutritional values of diet, since formulation is based on as-is fed basis.


Introduction
The storage and use of ingredients and diets for poultry is not always done in conditions that are favorable to their conservation. In order to improve the efficiency of the ingredients it is important to understand how they interact with the environment, especially in terms of water flow (Gabbi, Cypriano, & Piccinin, 2011).
The presence of water in diet occurs in two forms: in bound form, with restricted mobility, and in free form, which is water available for physical, chemical and microbiological reactions and subject to interaction with the environment, which is expressed by the water activity (aw), where the total moisture is determined by the sum of these two forms of water. This moisture is not static, since the ingredients adsorbs or desorbs moisture from the environment in which it is stored, depending on the temperature and relative air humidity. Research, Society andDevelopment, v. 9, n. 9, e828997729, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7729 4 Therefore, the water present in ingredients and diets is a risk factor for in-storage quality and conservation, since it can lead to deterioration by fungal development and mycotoxin production. Furthermore, it hinders handling and transportation and dilutes the total nutrients, proportionally reducing nutritional value of the product and consequently the animal performance and productivity (Coradi, P. C., 2010).
Understanding the interaction of ingredients with the environment and their hygroscopic behavior is of fundamental importance (Gabbi et al., 2011). Sorption isotherms represent important mathematical models for predicting product interaction with the environment, allowing one to determine the ideal humidity of the environment where the ingredient is stored in order to avoid the risk of microbiological growth and loss of nutritional value. Moreover, it is necessary to consider its moisture content for the precise dosage of ingredients used in poultry diet formulations (Park, Cornejo, & Fabbro, 2008).
The characteristics of sorption isotherms demonstrate a material's ability to absorb water from or release water into the surrounding environment, when placed in atmospheres with controlled relative humidity at a given temperature (Costa, C. M. L, 2010). Therefore, in order to provide information to support the adequate storage of ingredients and reduce risks of errors in diet formulations, this study aimed to determine the sorption isotherms of the ingredients commonly used in layer breeder diets.

Methodology
The experiment was conducted at the Physical-Chemical Analysis Laboratory of Embrapa Swine and Poultry, Concórdia/SC, and the samples were taken from ingredients at the Embrapa's animal feed mill.
During the experiment period the temperature (T°C) and relative humidity (RH%) of the medications storeroom and bagged ingredients storage area of the diet production facility were monitored using a data logger (Test Equipment -174 H), programmed to take Research, Society and Development, v. 9, n. 9, e828997729, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7729 5 measurements every 30 minutes. Weekly samples of three ingredients and diets (approximately 100 grams) were taken and stored in plastic containers.
A sample of each ingredient and of the animal diets were analyzed in as-sampled condition to determine the water activity (aw) and moisture content (M%) on an as-fed basis.
Eight samples of each material, weighing approximately five grams, were packaged in purpose-designed plastic capsules for the aw equipment and subjected to dehydration in a desiccator containing silica, in a vacuum, and at a temperature of 30°C for more than 24 hours in a model 002 CB oven. After that period, the samples were transferred to a desiccator containing water in the base and they were then placed into the oven again, and one sample of each material was removed at incremental intervals of 0, 1, 3, 5, 7, 8: 30; 24 and 26 hours from the start. Once removed from the oven, the sample was weighed on an analytical scale and then placed in the measurement chamber of the equipment for aw determination. Once the aw was determined, the samples were subjected to oven drying for 12 hours at 105°C and weighed again in order to ascertain the M%. The value was expressed on an as-fed basis and dry matter basis. This procedure was repeated three times for each ingredient sample. The RH% and T°C inside the desiccators were monitored using a data logger (Test Equipment -

H).
The aw and M% data for the samples subjected to humidification were evaluated using the GAB, BET, Halsey, Kunh, Henderson, Oswin, Mizrahi and Smith mathematical models, supported by the tool SOLVER (available on Microsoft Excel), establishing the isotherm for each of the ingredients. The GAB -Guggenheim-Anderson-de Boer mathematical model (non-linear regression analysisequation 01) was the only model to which the experimental data fitted in order to constitute the sorption isotherm. This model was chosen based on the highest determination coefficient value (R 2 ).

Results and Discussion
The moisture content and water activity data for the ingredient in sampled condition at the feed mill are presented in Table 1. It could be verified that the mash and pelleted poultry diets, corn, soybean meal, wheat bran, rice bran and BHT presented heightened moisture levels. The elevated relative humidity in the feed mill during the winter period might explain the heightened water activity and moisture content values (as-fed basis) of the ingredients and of the diets in sample condition (Table 2).

Temperature (°C) Summer Winter
Temperature ( Table 3 shows the adjustment constants of the GAB mathematical model (Xm, C and K) for the sorption isotherms.  According to Krist, Nichols and Ross (1999), ingredients with water activity values between 0.600 and 0.850 are considered intermediate, and those with values less than 0.600 are considered as having low water activity, therefore the data found fell into the intermediate water activity range (0.60 < aw < 0.85) and low water activity category (aw < 0.60). Lima and Sant'ana (2011), in a study with salted and dried fish, found intermediate water activity levels, varying between 0.74 to 0.75. Paglarini, Silva, Porto, Piasson and Santos (2013), working with isotherms of mango pulp (Mangifera indica L.) manteiga variety, and Moreira, Rocha, Afonso and Costa (2013), in a study with isotherms of mash lyophilized, also observed the best fit for of the experimental data by the GAB mathematical model.
Monolayer moisture (Xm) refers to the amount of water strongly adsorbed over the primary sites, considered as the unit where the ingredient is most stable. From the monolayer moisture one can ascertain the lowest water content contained in each ingredient. According to Goula, Karapantsios and Adamopoulos (2008), monolayer moisture is that which gives rise to (at a certain temperature) the greatest stability and minimal losses in the quality of the diet product. The Xm parameter is important as it can be related to the start of a series of diet deterioration chemical reactions (Ferreira & Pena, 2003). The increase in monolayer moisture associated to an increase in temperature is not common to all ingredients, and, according to Ferreira and Pena (2003), this behavior can be explained by the fact that a temperature Research, Society and Development, v. 9, n. 9, e828997729, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7729 increase can make modifications in the physical structure of the product, making a larger quantity of active sites with affinity to water molecules available or increasing the solubility of solutes in the product, thus retaining more water molecules in the monolayer.
The constants C and K from the GAB model are related to the effect of the temperature. According to Catelam, Trindade and Romero (2011), C is the constant that is related to the sorption heat in the monolayer and K refers to the sorption heat in the multilayer. According to Gabas et al. (2007), low temperatures favor the force of the adsorbate-adsorbent interaction, causing a certain increase in the values of the constant C. Timmermann (2003) states that the constant K of the GAB model increases with the force of the adsorbate-adsorbent interaction and values greater than 1 are physically inadequate, indicating an infinite adsorption.
In relation to determining the sorption isotherms of the ingredients used in formulating mash and pelleted diet for roosters and laying hens, a similar behavior was found in the format of the isotherms for all the materials. In accordance with the IUPAC classification (1985), the isotherms can present five different formats. As a rule, the experimental results of the sorption isotherms showed curves with sigmoidal behavior, implying type II isotherms, characteristic of soluble, protein and chalky products, and also common to food products (IUPAC, 1985).
According to Brunauer, Emmett and Teller (1938), these curves indicate the type of forces that intervene in the binding between the water and the hygroscopic material surface and enable evaluations of the superficial structure, stability during storage, and provide information for the design of adequate packaging to guarantee better product conservation. Therefore, as the formulation of poultry diet takes into account the ingredient in its asfed basis, special care should be taken with the moisture content of the ingredients, since there may be differences between the formulate and what is consumed by the animal, due to the water adsorption.
According to Rahman (2008) microorganisms begin to grow from a water activity level of 0.600 onwards. Hence, the ingredients and diets with water activity values of less than 0.600 are guaranteed against the growth of fungi, mold and bacteria.
It is found that at low water activity levels the ingredients adsorb small amounts of water, and at high levels of water activity there is an accentuated increase in adsorbed water, which result was also found by Cardoso, F. F. (2012), working with mixed flours of rice and barley husk.
Furthermore, the addition of water in the extrusion process, for example, results in increased moisture and water activity of diets, affecting the nutritional value of the dietary protein (Murakami, F. F., 2010).
Moreover, it is important to highlight that in their final portion, corresponding to the highest water activities, the isotherms demonstrated a more hygroscopic behavior, characterizing a rise in the curve. That same behavior was also verified by Canuto, Afonso and Costa (2014), when studying water adsorption of freeze-dried papaya pulp powder.
Slight decreases in the balance moisture are found with increase in temperature. Ferreira and Pena (2003) justified this behavior based on the increased vapor pressure of the water in the air and on the product surface, bearing in mind that this increase is greater on the product surface, as it contains more water molecules than in the air. Therefore, the greater vapor pressure results in a greater water loss so that balance is attained.
The sorption isotherms of the supplements are presented in Figure 1. In relation to the microbiological stability (aw < 0.6) of the supplements it is observed that the vitamin supplement will be stable when it presents a moisture content of less than 3.74%, the vitamin mineral supplement, 6.09% and the mineral supplement when the moisture is less than 3.00%.
For a given temperature, it can be verified that the moisture content of the product increases with water activity. According to Costa, C. M. L. (2010), this occurs because the vapor pressure of the water in the product accompanies the increased vapor pressure of the environment in which it is inserted. The sorption isotherms of the additives (Figure 1) used in the formulation of the diets demonstrate that the mycotoxin deactivator, choline chloride, kaolin, bacitracin zinc, limestone, phosphate, phytase, coccidiostat, salt and BHT will be microbiologically stable (aw < 0.6), when they present a moisture content of less than 6.52, 26.64, 1,02, 8.98, 0.06, 3.76, 9.55, 7.77, 0.33, and 26.51%, respectively. For the BHT, the data obtained did not fit for the determination of the sorption isotherm, in other words, it presented values for which it was not possible to observe any moisture adsorption by the ingredient.
In relation to the choline chloride, the ingredient with the highest hygroscopicity, care must be taken, since the increase in free water content in the mixture can lead to loss of hydrosoluble vitamins, as well as operational problems in the production of animal diets.

Final Considerations
The format of the sorption isotherms of the ingredients and diets assessed are type II, or sigmoidal, common among feed, protein and chalky products, except for BHT, which presented values that do not fit for isotherm determination.
Ingredients and diets have different hygroscopic behavior, which can lead to deterioration and low accuracy in nutritional values of diet, since formulation is based on as-is fed basis.