Chapter 5 - Physical and chemical changes in maize during processing

Contents - Previous - Next

Lime-treated maize (part I)

Chemical changes

The conversion of maize into tortillas involves a process in which water, heat and calcium hydroxide are used. All three influence the chemical composition of processed maize, causing changes in nutrient content. The changes that take place are caused by both physical losses of the kernel and chemical losses. The latter may result from destruction of some nutrients and chemical transformation of others.

The proximate composition of maize and of home-made and industrially prepared tortillas is shown in Table 16. Changes in fat and crude fibre content are shown, and in some cases an increase in ash content. The values for homemade and industrially produced tortillas are similar for most major chemical components with the exception of fat, which is higher in industrially produced tortillas.

Dry matter losses

From studies on maize cooking by rural housewives using their own traditional method, Bressani, Paz y Paz and Scrimshaw (1958) reported a loss of solids (17.1 percent for white maize and 15.4 percent for yellow maize) when maize was made into dough. Bedolla and Rooney (1982) have reported losses of 13.9 and 10 percent respectively for white and yellow maize using the traditional process and losses of 7 and 5.7 percent in steam cooking. In other studies where variations in the processing technique were evaluated, Khan et al. (1982) found losses of 7 to 9 percent in commercial processing, 9 to I I percent in pressure cooking and 11 to 13 percent using the traditional method. These workers also reported that dry matter loss increased as cooking time increased.

TABLE 16 - Proximate composition of raw maize and home-made and industrially produced tortillas

Sources: Bressani, Paz y Paz and Scrimshaw, 1958; Cravioto et al., 1945; Ranhotra, 1985; Saldana and Brown, 1984.

Likewise, the integrity of the maize kernel influences losses. Jackson et a/. (1988) reported that dry matter losses in the traditional cooking procedure were higher (10.8 to 12.1 percent) with broken kernels then with undamaged ones (6.3 to 8.9 percent). Besides the integrity of the kernel and the heating process used, other factors such as length of steeping influence dry matter losses. Long steeping caused larger losses than a short steeping time. Dry matter losses of QPM with a hard endosperm are similar to those of common maize. Recently, Bressani et al. (1990) reported losses of 17.1 percent for the Nutricta QPM variety as compared with 17.6 percent from a white tropical maize. Sproule e! al. (1988) found a 9.6 percent dry matter loss from QPM as compared with a 10.4 percent loss in common maize.

Dry matter losses depend, then, on a number of variables such as the type of maize (hard or soft endosperm), kernel integrity (whole or broken kernels), cooking procedure (traditional, steam cooking, pressure cooking, commercial), the levels of lime used, cooking time and steeping time, as well as other operations such as rubbing to eliminate the seed-coat during washing of the kernels. This process also eliminates other parts of the kernel: the tip cap and possibly the aleurone layer and small amounts of germ. Paredes-López and Saharopulus-Paredes (1983) used scanning electron microscopy to show that the outside surface of lime-treated maize had important structural deterioration. They indicated that the aleurone layer was retained as well as some pericarp layers and that the germ remained attached to the endosperm. Gómez et al. (1989) noted that important structural changes took place in maize during "nixtamalization". The alkali weakened the cell walls, facilitating the removal of the pericarp. It solubilized the cell wall in the peripheral endosperm, caused swelling and partial destruction of starch granules and modified the appearance of the protein bodies. The dough contained fragments of germ, pericarp, the aleurone and endosperm, as well as free starch and dissolved lipids. Thus some of the chemical changes that have been observed can be accounted for by the chemical compounds present in these three or four parts of the kernel. The dry matter content has been analysed by Pflugfelder, Rooney and Waniska (1988a), who reported 64 percent non-starch polysaccharides (fibre), 20 percent starch and 1.4 percent protein.

Nutrient losses

Studies on the losses of nutrients during the transformation of maize into tortillas are not abundant, even though significant changes due to processing do take place (Cravioto et al., 1945; Bressani, Paz y Paz and Scrimshaw, 1958). Ether-extractable substances are lost, 33 percent in yellow maize and 43 percent in white maize. This is difficult to explain, although it could be partially accounted for by the loss of the pericarp, the aleurone layer, the tip cap and some of the germ, parts of the kernel containing ether-extractable substances. Losses in crude fibre were reported to be about 46 percent in white maize and 31 percent in yellow maize. Lime treatment at 96°C for about 55 minutes hydrolyses the pericarp, which is removed during washing, pulling the tip cap with it, and this would account to a large extent for fibre loss. Nitrogen losses amount to about 10 and 5 percent for white and yellow maize, respectively. Again, this may be partly due to the physical loss of the pericarp and tip cap. Even though tortillas may have a slightly higher protein content than the original maize on an equal moisture basis, as has been reported by various workers, this may be caused by a concentration effect, since soluble sugars from the kernel are lost. Ash content increases because of the absorption of lime, which significantly increases calcium content (Saldana & Brown, 1984; Ranhotra, 1985). Significant losses take place in thiamine (52 to 72 percent), riboflavin (28 to 54 percent) and niacin (28 to 36 percent). In yellow maize 15 to 28 percent of the carotene was lost (Cravioto et al., 1945; Bressani, Paz y Paz and Scrimshaw, 1958).

Fat and fatty acids. Ether-extractable substances of 33 and 43 percent were reported by Bressani, Paz y Paz and Scrimshaw (1958) from yellow and white maize respectively, as processed in Guatemalan rural homes. Pflugfelder, Rooney and Waniska (1988b) found losses of 11.8 to 18.1 percent and suggested that these could be partly due to the vigorous handling of cooked maize at the industrial plant. Of the total masa lipid, 25 to 50 percent was free and partially emulsified. Bedolla et al. (1983) found ether extract values of 5.0, 3.1 and 3.6 percent in raw maize, cooked maize and tortillas respectively, or about a 28 percent change. This loss has not been fully explained; however, it may result from the loss of the seed-coat, the tip cap, the aleurone layer and possibly part of the germ, and also from ether soluble substances, not necessarily fat. Even though ether-extractable substances are lost in the process of converting maize into tortillas, the fatty acid make-up of the fat does not change in common maize or QPM, as shown in Table 17. Differences between maize samples, either raw or processed, are larger than those between raw maize and tortillas, suggesting that the alkaline cooking method does not alter the fatty acid make-up of the fat.

TABLE 17 - Fatty acid content of common and quality protein maize and tortillas (%)

Source: Bressani et al., 1990

Fibre content. The crude fibre content of maize - as determined by the Association of Official Analytical Chemists (AOAC) methodology decreases as the kernel is converted into tortillas. Various investigators (e.g. Saldana and Brown, 1984) have explained how and why such a loss takes place. With newer methodology to determine fibre, Reinhold and Garcia (1979), using the Van Soest method, reported that the neutral detergent fibre (NDF) and acid detergent fibre (ADF) in tortillas (6.60 and 3.75 percent, respectively, on a dry weight basis) were significantly higher than those found in the dough (an average of 5.97 and 2.98 percent, respectively). No difference was reported in hemicellulose, with dough containing 3.18 percent and the tortillas 2.89 percent. Using the same method, Bressani, Breuner and Ortiz (1989) found 10.8 percent NDF in maize and 9 percent in tortillas, as well as ADF of 2.79 and 3 percent respectively. Hemicellulose averaged 8 percent in maize and 6 percent in tortillas, while the values for lignin were 0.13 and 0.15 percent. These values and others are shown in Table 18. Using the method of Asp et al. (1983), Acevedo and Bressani (1990) detected a decrease in insoluble fibre from raw maize (13 percent) to the dough (6 percent) and an increase in tortillas (7 percent). Soluble fibre increased from 0.88 percent in raw maize to 1.31 percent in the dough, and further increased to 1.74 percent in tortillas. Fibre decreases from raw maize to dough are due to the losses in seed-coat described previously. Increases from dough to tortillas, however, may be due to the browning reaction, as has been reported in baked wheat products (Ranhotra and Gelroth, 1988).

TABLE 18 - Dietary fibre in common and quality protein make and tortillas (%)

Sources: Acevedo and Bressani, 1990; Bressani, Breuner and Ortiz, 1989; Bressani et al., 1990; Krause, 1988; Ranhotra, 1985; Reinhold and Garcia 1979.

Ash. Changes in ash content have not received much attention from researchers. Most findings, however, have shown an increase in total ash content from maize to tortillas, which may be expected because of the lime used for cooking. Along with this increase in ash there is a significant increase in calcium content. According to Pflugfelder, Rooney and Waniska (1988b), calcium content in the dough is influenced by lime levels, cooking and steeping temperatures and maize characteristics. The changes in other minerals are variable and may depend on the purity of the lime used as well as on the type of grinding equipment. In one study (Bressani, Breuner and Ortiz, 1989; Bressani et al., 1990) the magnesium content increased from 8 to 35 percent from maize to tortilla; there was no change in sodium and a small decrease in potassium. Iron content also increased; however, the increases may have resulted from contamination. Phosphorus content also increases from maize to tortilla (Table 19). One aspect of nutritional interest is that the calcium-to-phosphorus ratio, which is about 1:20 in maize, changes to approximately 1:1 in the tortilla.

TABLE 19 - Mineral content of raw maize and home and industrial samples of tortillas (mg/100 9)

Sources: Bressani et al., 1990; Krause, 1988; Ranhotra, 1985; Vargas, Munoz and Gómez 1986

Carbohydrates. Maize and tortillas contain significant amounts of soluble carbohydrates, but very little is known on how they change during alkaline processing. Starch losses of about 5 percent have been reported; these are recovered in the solids lost. A decrease in sugar from 2.4 percent in maize to 0.34 percent in tortillas was also found. Robles, Murray and Paredes-Lopez (1988) found that alkali-cooking and soaking of maize caused large increases in viscosity and that cooking time had a significant effect on pasting properties, although there was no extensive gelatinization of the starch. Differential scanning calorimetric studies yielded similar gelatinization endotherms for untreated maize and nixtamal flours. In the process enzyme-susceptible starch increases as cooking time lengthens.

Protein and amino acids. Most researchers report a small increase in N content which is attributed to a concentration effect. The solubility of all protein fractions is decreased from raw maize to tortillas, with an increase in the insoluble fraction.

Bressani and Scrimshaw (1958) extracted the nitrogen from raw maize and tortillas using water, sodium chloride, 70 percent alcohol and sodium hydroxide. The solubility of the water, salt and alcohol protein fractions was significantly lower in tortillas, with the alcohol-soluble proteins affected most. Only a small decrease of about 13 percent in the solubility of the alkali-soluble fraction was detected. Because of this, the insoluble nitrogen fraction increased from 9.4 percent in maize to 61.7 percent in tortillas.

Ortega, Villegas and Vasal (1986) observed similar changes in both common and QPM maize using the Landry-Moureaux (1970) protein fractionation technique. The solubility of true zeins decreased 58 percent in the tortillas prepared from common maize and 52 percent in QPM tortillas. The authors indicated that hydrophobic interactions may have been involved in the change in protein solubility observed. Sproule et al. (1988) noted a decrease in the albumin plus globulin-nitrogen, expressed as percentage of total nitrogen, from maize to tortillas.

The changes in amino acid content from maize to tortillas are summarized in Table 20. In vitro enzymatic studies of amino acids indicated that total nitrogen and alpha-amino nitrogen were released faster from maize than from tortillas. Values for alpha-amino nitrogen released, expressed as a percentage of the total nitrogen release, were higher for tortillas than for raw maize after 12 hours of hydrolysis with pepsin. The percentage of alpha-amino N from the total was similar for maize and tortillas at 60 hours of hydrolysis with trypsin and pancreatic. After 60 hours of hydrolysis with pepsin, trypsin and pancreatin, the percentage of enzymatically released amino acids with respect to the acidhydrolysed amino acids suggested a faster release from tortillas than from maize. This information was recorded up to 36 hours for most of the amino acids except leucine, phenylalanine, tryptophan and valine, which were released at about the same rate. At 60 hours of hydrolysis the amino acid concentrations of the maize and tortilla hydrolysates reached comparable levels, except for methionine (Bressani and Scrimshaw, 1958). These authors reported losses of arginine (18.7 percent), histidine (11.7 percent), lysine (5.3 percent), leucine (21 percent), cystine (12.5 percent) and small amounts of glutamic acid, proline and serine.

Sanderson et al. (1978) found small losses of arginine and cystine from alkaline treatment of common and high-lysine maize. These same authors found 0.059 and 0.049 g of Iysino-alanine per 100 g protein from common and high-lysine maize respectively, but none was found in raw maize. In commercial masa, they found 0.020 g Iysino-alanine per 100 g protein, while in tortillas the level found was 0.081 g per 100 g protein.

TABLE 20 - Amino acid changes during the alkaline cooking of maize (9/16 g N)

Sources: Bressani and Scrimshaw, 1458; Sanderson el al., 1978

Lunven (1968), using his own amino acid column chromatography technique, observed significant losses in both lysine and tryptophan during the alkaline treatment of common maize. Ortega, Villegas and Vasal (1986) found small losses in tryptophan in tortillas of both common maize (about 1 1 percent) and QPM (15 percent). On the other hand, they reported minimal losses in lysine from both types of maize, similar to those previously noted. Higher losses for both amino acids have recently been reported by Bressani et al. (1990) from common maize and QPM (Nutricta) maize converted into tortillas by rural processing. Ortega, Villegas and Vasal (1986) also indicated that on the basis of the very small loss of lysine in the alkaline product, minimal amounts of lysino-alanine were probably present in the tortillas of common maize and QPM used in their study.

TABLE 21 - Vitamin content of raw maize and tortillas (mg/100 9)

Sources: Bressani, Paz y Paz and Scrimshaw, 1958; Cravioto e, al., 1945; Ranhotra, 1984; Saldana and Brown, 1984.

Vitamins. Losses in thiamine, riboflavin, niacin and carotene occurred during processing of maize into tortillas by lime-cooking. A summary of some data is shown in Table 21. The vitamin that has attracted the attention of a number of researchers has been niacin because of its relationship to pellagra. The biological implications of the lime-cooking process on niacin availability and pellagra is discussed in the next section. This section discusses the changes in concentration of niacin resulting from limecooking. Bressani, Gómez-Brenes and Scrimshaw (1961) reported that the seed-coat of maize contained 4.2 mg niacin per 100 g, while the germ and endosperm contained about 2 mg niacin per 100 g. About 79.5 percent of the kernel niacin was provided by the endosperm, and 10 percent each by the germ and seed-coat. After lime-cooking, the endosperm contributed about 68 percent of the total niacin and the germ about 5.5 percent. After cooking, 26 percent of the total was found in the cooking water. The percentage of niacin extracted in water was 68.5 percent of the total with raw grain and 76 percent with lime-cooked maize. Furthermore, enzymatic hydrolysis with pepsin yielded 69 percent of the niacin of all the samples, and after trypsin and pancreatin hydrolysis niacin yields were 78 and 100 percent respectively. This information was interpreted to mean that niacin is slightly more available from lime-treated maize than from raw maize.

Contents - Previous - Next