Criteria of Determination of Safe Grain Storage Time – A Review

contents and temperatures and with lower levels of mechanical

Knows in the bibliography of the subject are tables and graphs of the storage times. Sometimes, however, mathematical formulas are more useful. They can be easily incorporated into the mathematical models of grain drying or aeration and expert systems which are aids for storage-grain management (Arinze et al., 1993;Courtois, 1995;Fleurat-Lessard, 2002;Kaleta, 1996). Such formulas known in the bibliography of the subject and own formulas developed by the authors of the chapter are presented in the paper.
To test the effect of grain parameters on the safe storage period, three criteria have been applied: carbon dioxide (CO 2 ) production and dry matter loss, appearance of visible moulds, and germination capacity.

Carbon dioxide production and dry matter loss
Grain deterioration is related to respiration of the grain itself and of the accompanying microorganisms. Respiration is the process of oxidizing (combusting) carbohydrates and yielding carbon dioxide, water vapour and energy. Therefore, respiration consumes dry matter.
The complete combustion (aerobic respiration) of a typical carbohydrate such as starch is represented by the following equation: 6 According to this equation during the breakdown of 1 g of dry matter by aerobic respiration using 1.07 g of oxygen, 1.47 g of carbon dioxide, 0.6 g of water, and 15.4 kJ of heat energy are released. It means that a 1 % loss in grain dry matter carbohydrate is accompanied by the evolution of 14.7 g of carbon dioxide per kg of grain matter. Therefore respiration rate is closely related to grain dry matter loss and, consequently, global quality loss. Modelling CO 2 production can be used to simplify the prediction of rate of quality loss, assuming predominantly aerobic respiration.
Contamination of harvested grain by microorganisms is natural and permanent. In temperate climates with medium wet or moist grain at harvest, the genera Fusarium, Alternaria and Helmintosporium (called "field flora") are predominant. During long term storage, xerophilic fungi of the genera Aspergillus and Penicillium (called "storage flora") progressively replace the "field flora" over a period of several months of storage. At 15-19 % moisture content, most species of the field flora are inhibited or die whereas storage flora species slowly grow (Fleurat-Lessard, 2002;Frisvad, 1995;Pelhate, 1988). Since the respiratory processes of microorganisms or of hidden insect infestation are similar to those of the grain itself, the combustion of carbohydrates is a representation of grain, microorganisms and insect respiration (Fleurat-Lessard, 2002; Sinha et al., 1986;Steele et al., 1969).
The following mathematical formulas for predicting carbon dioxide production and dry matter loss can be found in the bibliography of the subject. White et al. (1982) carried out numerous experiments on the carbon dioxide release rates of wheat and developed the following equation for general prediction of the instant rate of CO 2 release from grain: 2 w X a bT ct dt eM = + + + + (2) where X is the rate of CO 2 production in mg kg -1 d. m. per 24-h period, T is the grain temperature in °C, t is the time in h, M w is the grain moisture content in % w. b., and a, b, c, d, and e are empirical constants.
The following equation was developed by Srour (1988): where X is the rate of CO 2 production in mg (100 gd. m.) -1 per 24-h period, M w is the grain moisture content in % w. b., T is the grain temperature in °C, and a, b, and c are empirical constants.
whereX is the rate of CO 2 production in mg d -1 kg -1 d. m., t is the storage time in d, and M w is the moisture content in % w. b.
There are however some problems in using equations (2) -(4) to describe quality changes. They are based on the measurement of CO 2 release rate, either from a grain sample or directly in a grain bin. Such measurements can be done using sophisticated equipment and in laboratory conditions. Additionally, when grain moisture is below 14 % (w. b.) the release rate is very low and therefore it is very difficult to measure it. However, the above formulas are not useful in prediction the storage life.
Another option for the prediction of safe storage life is the calculation of dry matter loss (DML) as a function of grain temperature, grain moisture content, and storage time. Seib et al. (1980) stated that the amount of dry matter loss from respiration is an indication of grain quality. They also stated that rough rice stored at 15 % and 18 % w. b. moisture content fell below U. S. Grade Nos. 1 and 2 if DML exceed 0.75 %. Some authors assumed that an acceptable level of dry matter loss is 0.5 %. In high moisture maize (corn, 25 % m.c.) a loss of 0.5 % dry matter can occur in 7 days, sometime without any visible moulding. However, this way found to be sufficient to render maize grain unfit for use, and also to produce aflatoxins (Marin et al., 1999). Kreyger (1972) considered grain to be fit for animal feed with DML of up to <2 %. However, Hall and Dean (1978) suggested 1 % DML was acceptable in grain for food use and that this could be applied to both wheat and maize. White et al. (1982) stated that 0.1 % was unacceptable for wheat of premium grade and proposed the absolute limit of DML at 0.04 %. Therefore the problem of what is the limit for an acceptable level of dry matter loss is still controversial. Seib et al. (1980) developed the following expression to determine DML of long-grain rough rice as a function of grain temperature, grain moisture content, and storage time: where DML is the dry matter loss in decimal form, t is the storage time in h 10 -3 , T is the grain temperature in °C, M w is the grain moisture content in decimal w. b., and A, C, D, and E are empirical constants.
Equation (5) where x=At C , y=D (1.8 T-28), z=E(M w -0.14) The values of the constants for long-grain rough rice used in the equations (5) and (6) where DML is the monthly dry matter loss in %, T is the grain temperature in °C, M w is the grain moisture content in % w. b., and 5°C≤T≤20°C, 14 %w. b. ≤M w ≤35 % w. b.
From equation (5) and from Scherer (1980) data increase in the dry matter loss with the increase of both grain temperature and moisture content can be observed. In such conditions the respiration of grains is more intensive. DML increase with the duration of the grain storage.
Scherer's et al. (1980) investigations on damaged grain confirmed the negative influence of mechanical damages on dry matter loss shown by Thompson (1972). Scherer et al. (1980) stated that increase in amount of damaged corn caused the decrease in safe storage time.
They accepted the limit of DML at 0.5 % and observed that 1 % of damaged grain together with 1 % of chaff and fines reduced the safe storage time in almost 6 %, however 20 % of damaged grain and 5 % of chaff and fines reduced the time in almost 38 %. They explained obtained results by more intensive respiration of chaff and fines, and damaged grain comparing with undamaged grain.
where t is the storage time in d, T is the grain temperature in °C, M w is the grain moisture content in % w. b., and 1°C≤T≤24°C, 15 %w. b. ≤M w ≤30 % w. b.
Al-Yahya (2001) examined the conditions of safe storage of wheat. Based on these data the following relationship between storage time, grain temperature, grain moisture content and DML can be presented: where t is the storage time in d, T is the grain temperature in °C, M w is the grain moisture content in % w. b., DML is the dry matter loss in %, and 4°C≤T≤40°C, 15 According to equation (1) heat energy is released during the respiratory process of grain, microorganisms and insects. The heat produced within the pockets of wet grain is especially harmful. It is not dissipated rapidly because of the low thermal conductivity of the grain (Kaleta, 1999; Kaleta and Górnicki, 2011) and the slow free convection currents in the granular bulk. The elevated grain temperature and moisture content of the pocked provide a favourable environment for further growth of microorganisms, thereby making the heating process self-accelerating. Heat production in stored grain ecosystems was investigated by e.  Wilson (1999) proposed a mathematical model for predicting mould growth and subsequent heat generation in bulk stored grain. Unlike previous models, it was intended to be applicable in conditions that change with time. Starting from a model for mould growth in varying conditions the work of a number of authors was combined to produce a model to predict the heat production at all parts in a grain bulk. The effect of temperature and relative humidity on the mould growth rate was decoupled, so that the resulting equation for mould growth was a product of one-parameter terms. The heat generation rate was then written as a specific function of the mould population and mould grow rate. The model's current predictions for very wet grains was good, but for dried grain model performs less well.

Appearance of visible moulds
Spoilage of grains is the result of microorganisms (bacteria, yeast, fungi, and moulds) utilizing the nutrients present in the grain for growth and reproductive processes, spoilage may result in a loss of nutrients from the grain since microorganisms use these nutrients in much the same way as livestock. Also, microorganisms produce heat and moisture during growth which can cause a temperature rise in stored grain. Heating initiated by microbial growth can cause "heat damage" and can sometimes render grain unfit for feed. Such conditions have been known to cause fires and dust explosions in storage structures (Ross et al., 1979).
Certain microorganisms, when allowed to grow under the proper environmental conditions, can produce toxins or other products which, if consumed by either livestock or humans, can cause serious illness and even death. A number of these toxins and the microorganisms which produce them have been identified.
Toxigenic fungi infect agricultural crops both in the field and in storage. Converse et al. (1973) found the following variety of fungi in the corn at harvest and after 28  Fungal infections can be discolour grain, change its chemical and nutritional characteristics, reduce germination and, most importantly, contaminate it with mycotoxins, the poisonous metabolites produced by certain fungal genera.
Ergot is a disease of cereal crops caused by the fungus Claviceps purpurea. It causes reduced yield and quality of grain. The effect of the ergot's alkaloid toxins on man and animals is, however, of much greater significance (Moreda and Ruiz-Altisent, 2011).
Aflatoxins are secondary metabolites produced by Aspergillus flavus Link and A. parasiticus Mites also infect stored cereals. These arthropods contaminate grains and are a matter of great concern in the medical and veterinary fields, since they may act as carriers of bacteria and toxigenic fungi. Grains contaminated by mites may cause acute enteritis when ingested, and severe dermatitis and allergy in cereal handlers. Furthermore, mites can feed on the germ of kernels, thereby reducing the content of iron and vitamins of the B complex and germination ability. Stored -product mites can survive and multiply by feeding on several species of seed -borne fungi. Fungal spores and mycelia contain small amounts of essential nutrients (e. g. vitamins of the B complex and steroids), and moisture levels adequate for the metabolic demands of mites. The constant migration of mite populations within a granary ecosystem efficiently contributes to the dispersal of viable fungal spores of several species, including Aspergillus spp. and Penicillium spp., carried on the vector's body surface or deposited with its feces (Franzolin et al., 1999).
Conditions favouring the development of mycotoxins in cereals before and after harvest are important to grain -exporting countries concerned with marketing high -quality products.  There is, however, lack of simple equations predicting the length of safe storage period by a combination of, at least, moisture content of grain and storage temperature.
where t is the storage time in h, T is the grain temperature in °C, M w is the grain moisture content in % w. b., and 5°C≤T≤25°C, 16 % w. b. ≤ M w ≤26 % w. b. Kreyger (1972) investigated the safe storage times of several grains. He assumed that the best criterion for safe storage times is the one that is based on the time to the appearance of visible moulds. Based on Kreyger's (1972) data, we developed the following formula: where t is the storage time in weeks, T is the grain temperature in °C, M w is the grain moisture content in % w. b., A, B, C are empirical constants given in Table 1, and 10°C≤T≤25°C. Equation (17) and (18) confirm that the duration of the safe storage time increases with the decrease of both grain temperature and grain moisture content. Such conditions are not favourable for the mould development.
There are, however, controversies about the criterion of appearance of visible moulds. Several researches (Ryniecki and Nellist, 1991;Nellist, 1998), followed Kreyger (1972), took it as the best criterion for safe storage time. Some of them (Armitage, 1986;Fleurat-Lessard, 2002) mentioned, however, several drawbacks of this criterion. The main drawback of this kind of prediction of safe storage life of stored grain is the subjective determination of visible mould on the kernel. Another drawback is the lack of progressiveness in the prediction. Before the onset of visible spoilage, grain is theoretically sound and its quality is not altered. The day after spoilage is seen, the grain is deteriorated and should be downgraded.

Germination capacity
Various factors can reduce the storage life of some premium grade quality cereals. Moisture content of the harvested grains and storage temperature can encourage mould and insect pest damage. The best studied quality parameter is germination capacity, which is only of direct importance for grains. Nevertheless, this is probably the best surrogate measure of cereal grain soundness (Pomeranz, 1982). Cereals retaining a high level of viability in storage are also likely to retain the other main parameters of commercial or technological quality (Fleurat-Lessard, 2002).
Germination is defined as the appearance of the first signs of growth, i. e. the visible protrusion of the radical (Black, 1970). Germination can be affected by many factors such as grain temperature, grain moisture content, grain damages, fungus and insect infection. Much research has been conducted to determine the effect of various factors on germination.
McNeal (1966) found that soybean can be kept for 12 months without an expressive decline in germination if the temperature is kept below 16°C and the moisture content is not higher than 16.2 %, dry basis. Mayeux et al. (1972) noted than the germination of soybean seed is influenced by the percentage of split beans in stored seed, and storage temperature and moisture play an important role in maintaining the soybean seed quality. Kreyger (1972) used percentage germination as an indicator of grain deterioration. He studied the effect of many levels of grain moisture content and grain temperature on the percentage germination. His findings will be discussed below. Parde et al. (2002) studied the storage behaviour of soybean seed and the loss in quality due to free-fall from different heights (0.5-2 m) on to different surfaces (cement and galvanized iron) were studied. They found that soybean seed is susceptible to mechanical damage. The severity of damage varies with moisture content of seed because the dryer seed is harder. The hight of fall produces significant effects on germination. An average germination loss of 10 % and 31 % was noticed when the seed fell from a height of 1 and 2 m, respectively, on to the cement floor. This drop in germination was 7.5 % and 22 % when dropped from the same heights on to galvanized floor. The seed lots held at 12 % moisture content, dry basis, suffered less damage during free-fall from different heights than the lots held at 10 % and 11 % m.c. Soybean seed lots at 12 % m.c. retained germination ability for a longer period than the seed lots at lower m.c. There is, however, lack of simple equations predicting the length of safe storage period by combination of, at least, moisture content of grain and storage temperature.
where t is the storage time in d, M w is the grain moisture content in % w. b., T is the grain temperature in °C, and 10°C≤T≤40°C. Kaleta (1996) used equation (19) in her computer program developed to simulate wheat drying in silos with radial (horizontal) and vertical airflow, predict grain spoilage under the si-mulated conditions, and determine the most advantageous conditions of conducting the process of wheat drying in silos.
Muir and Sinha (1986) developed a set of two regression equations for predicting allowable storage times for canola before the germination capacity drops by 5 %.
where t is the storage time in d, M w is the grain moisture content in % w. b., T is the grain temperature in °C, and 10°C≤T≤40°C.  (21) where n is the number of simulated time steps. SI is a spoilage or storage index and its instantaneous value represents the progress of grain spoilage. A spoilage index of 1 or greater indicates that the allowable storage time has elapsed and the 5 % loss in germination has occurred to the canola.
where t is the storage time in d, and T is the grain temperature in °C. They stated also, that the safe storage times of 17 % m.c. wheat were 5, 7, and 15 d at 35,30, and 25°C, respectively.
The germination capacity of wheat at 17-19 % m.c., wet basis, stored at 25°C can be predicted from the measured respiration rate and moisture content by the equation (Karunakaran et al., 2001): where Y is the germination capacity of grain in %, X is the rate of CO 2 production in mg d -1 kg -1 d. m., and M w is the grain moisture content in % w. b.
Equation (23) where Y is the germination capacity of grain in %, t is storage time in d, M w is the grain moisture content in % w. b., and X is the rate of CO 2 production in mg d -1 kg -1 d. m.
Based on the germination data of Kreyger (1972), we developed the following formulas for predicting allowable storage times: were t is storage time in weeks, T is the grain temperature in °C, M w is the grain moisture content in % w. b., A, B, C are empirical constants given in Table 2, and 10°C≤T≤20°C.
were Y is the germination capacity of grain in %, M w is the grain moisture content in % w. b., MD is the mechanical damage in %, A, B, C, D, E, and F are empirical constants given in Table 3, and 15 %w. b. ≤M w ≤24 %w. b., 0≤MD≤30 %.  Table 3. Values of coefficients in equation (26) and the second one where Y is the germination capacity of grain in %, T is the grain temperature in °C, MD is the mechanical damage in %, A, B, C, D, E, and F are empirical constants given in Table 4, and 4°C≤T≤40°C, 0≤MD≤30 %.
Equations presented in this section confirm that the changes in germination capacity of stored grain are lower with lower following parameters: grain temperature, grain moisture content, mechanical damage and storage time. In general, the conclusions are the same as in previous section: longer storage times are possible with lower both grain moisture contents and temperatures and with lower levels of mechanical grain damages.
At the end of the chapter it is worth to mention shortly the other grain quality criteria which can be important to consumer and food manufacturer.
Colour of white rice is an important criterion for judging quality and price. The white colour becomes yellow after a period of storage. Dry matter loss of grain and heat liberated from its respiration and biological activities may accelerate rice yellowing. Parameters affecting the rice yellowing are temperature and relative humidity (water activity) ( Table 4. Values of coefficients in equation (27) Corn quality can mean wet-milling quality. It corresponds to the amount of survival thermo-sensitive proteins inside the grains and is very well correlated with the thermal history of the grains (Courtois, 1995).
The rate of quality changing can be represented with a simple zero-or first-order reaction (Labuza, 1980): where A is amount of a quality factor, ±dA/dt is the rate loss of a quality factor or production of undesirable effects, k 0 is the pre-exponential factor, E A is the activation energy in J mol -1 , R is the gas constant in J mol -1 K -1 , T is the temperature in K, and n is the reaction order (1 for first-order, 0 for zero-order).
where b is yellowness of rice in Hunter b unit, t is the time in d, k is the constant value for the yellowing rate in Hunter b unit d -1 , RH is the relative humidity in decimal, T is the temperature in K, and 308 K≤T≤338 K, 0.80≤RH≤0.95. Courtois (1995) where Q is the wet-milling quality, t is the time in s, k 0 is the pre-exponential factor in s -1 , T is the temperature in K, M is the grain moisture content in decimal d. b., and R is the gas constant in J mol -1 K -1 , E A =-133200 J mol -1 .

Conclusion
Nowadays grain is harvested with a combine harvester. Therefore it is possible to delay the process and to harvest ripe and dry grain, without any bigger losses caused by ridging of grain, yet in certain parts polluted with green parts of plants, straws and seeds of weeds or with unripe caryopses, moisture content can even exceed 30% w. b., and temperature is often above 30°C. This state can cause self-heating processes even when the grain itself is considered as dry. In such grain and even in grain considered as dry, vital functions connected with metabolism still exist, namely grain respiration, growth of moulds and other microorganisms as well as growth of insects. These processes lead to a decline in the quality of grain and even to its entire damage. The intensity of these processes depends mainly on the moisture content of grain and its temperature. For the purpose of safe grain storage one ought to limit its vital functions as soon as possible through lowering moisture content and temperature reduction. It can be realized by drying, and then cooling the grain. Due to economy in thermal energy consumption, grain is often dried with the atmospheric air or slightly heated air, but such a process runs very slowly, and grain has to stay in the drying chamber for quite a long time. During harvest, when granaries accept large quantities of harvested grain, it is not always possible to immediately clean, dry and cool the grain due to the limited capacity of devices. Therefore there is a necessity of periodic storage of the fresh grain mass, so there is a risk that undesirable processes will occur, which can lead to a decline in quality, and even entire damage of grain. It is therefore necessary to determine the time of safe grain storage, i. e. the time in which the growth of undesirable processes does not cause any essential changes in the quality of grain. The basic criteria of determination the length of this period are: CO 2 production and connected with it loss of the dry matter of grain, appearance of visible moulds, and germination capacity.
The dependencies for determining the time of safe grain storage were discussed. The general conclusions for all discussed criteria are the same: longer storage times are possible with lower both grain moisture contents and temperatures and with lower levels of mechanical grain damages.

Agnieszka Kaleta and Krzysztof Górnicki
Faculty of Production Engineering, Warsaw University of Life Sciences, Poland