The Effect of Low Levels of Antioxidants on the Swelling and Solubility of Cassava Starch
D.B Mat Hashim, S.N. Moorthy (Trivandrum, India)
J.R. Mitchell, S.E. Hill, K. J. Linfoot, and J. M. V. Blanshard, Loughborough, Leics. (United Kingdom)
The addition of low levels of sodium sulphite had a dramatic effect on the structure of cassava (Manihot esculenta Crantz) starch pasted at 95oC. Inclusion of 100 ppm sulphite promoted granule disintegration resulting in a reduction in measured swelling volume to nearly 10% of its value in the absence of sulphite and release of almost all the carbohydrates from the starch granukle. It is suggested that sulphite addition promotes oxidative reductive depolymerization (ORD) of the polysaccharides since its effect on the granules is prevented when lower levels (around 20 ppm) of the polar antioxidant propyl gallate are added. Possible practical applications of the results and implications for the understanding of starch granule structure are discussed.
The changes that occur to starch on heating have been extensively studied. When starch is heated in excess amounts of water above its gelatinisation temperature, the granule swells and carbohydrate is released from the swollen granule. These effects can be quantified by centrifuging the suspension thereby obtaining a swelling index from the weight of the sedimented swollen granules while the released components may be obtained gravimetrically from the soluble solids content  or the carbohydrate content of the supernatant . The extent of swelling and the composition of the material that leaves the starch granule depend upon the botanical source of the starch  as well as the heating and shearing procedure .Starches show different patterns of swelling and solubility when heated at different temperatures. Cereal starches show a two stage swelling and solubility pattern [3,4,5]. In the region of the gelatinisation temperature, there is limited swelling and only a small amount of carbohydrate is solubilised. At around 90oC there is a large increase in the extent of swelling and a more substantial loss of carbohydrate from the granule. Normally, amylose leaves the granule first and it is only at the higher temperature stage of the pasting process that the amylopectin is lost , although for oat starch it has been shown that both amylose and amlopectin are leached out together but only when held in the 90 to 100oC temperature range . In contrast to cereal starches, potato starch shows massive single stage swelling near the gelatinization temperature although high carbohydrate leaching only occurs at around 900C. There seems to be some disagreement in the literature on the swelling properties of casaava (Manihot esculenta Crantz) starch. Results presented by Som et al.  and Rasper  showed two stage swelling whereas swelling patterns presented by Leach et al  and Kawabata et al  showed a single stage process. The swelling index and solubility are dependent on the variety and the age of the crop at which the starch is extracted . The physical and chemical properties of cassava starch have been recently reviewed by Rickard et al. .
In addition to the microscopic processes of gelatinization and swelling, heat will also induce deploymerization of amylose and amyloprotein. It is well recognized that the rate of such depolymerization when investigated for an isolated polysaccharide, is substantially reduced when oxygen is removed from the solution . This suggests that oxidative reductive depolymerization (ORD) reactions are important. There has been substantial recent interest in minimizing these reactions by the use of antioxidants. For example, there has been extensive work on xanthan gum because of the importance of heat stability in oil field applications . More recently, it was demonstrated that the use of the binary antioxidant system of sodium sulphite and propyl gallate was effective in preventing loss in viscosity and gel strength in retorted galactomannan systems [14, 15]
It seems possible that a significant contribution to swelling ans subsequent disintegration of the granule at high temperatures is the breakage of some linkages particulary in the amyloprotein component, which facilitate the loss of this material from the swollen granule structure. If this id due to ORD reactions, it may be expected this process could be controlled by the use of the binary antioxidant system which has been employed successfully with galactomannans. This idea prompted us to investigate the effect of sodium sulphite and propyl gallate and their combinations on the swelling volume and carbohydrate release from cassava starch pasted at high temperature.
Materials and Methods
Cassava starch (M4 variety) was obtained from the Central Tuber Crop Research (CTCRI), Trivandrum, India. The moisture content (wet basis) was 10.4%  and the amylose content was reported to be about 20.9% . Sodium sulphite (AR grade) was obtained from Fisons PLC and n-propyl gallate from Sigma Chemical Company.
Preparation of dispersions
Solution of different concentrations of sodium sulphite
Measurement of swelling volume
The swelling volume was determined by heating 15 ml of an aqueous 1% (w/v) starch suspension in screw top 40 ml Universal sample bottles. The samples were heated in a 95oC water bath with gentle shaking to ensure the granules remained in suspension until gelatinisation occurred and then left in the water bath for a further period of one h. after cooling, the samples were transferred into 15 ml conical centrifuge tubes and centrifuged (using a swing out CENTAUR 2)at 2200 rpm (approx. 1 000 x g) for 20 min. The swelling volume was obtained directly by reading the volume of sediment in the tube. The swelling volume is the volume of sediment per 100 ml of starch sample.
Total solids and carbohydrate content
The total solids content in the supernatant was measured by drying about 3 g of the supernatant in an air oven at 105oC for 24 h. The samples were then placed in a desiccator to cool for 3-4 h prior to weighing. The total carbohydrate content of the supernatant was determined after dilution using an autoanalyser with 70% sulphuric acid containing 1% orcinol as reagent. The samples were measured against glucose standards.
The absorbance of the supernatant was measured at 330nm using a spectrophotometer (LXB Biochrom Ultrospec 4050), with distilled water as the reference sample.
Effect of sodium sulphite
Fig 1. displays the effect of sulphites alone on the swollen volume. There is a pronounced minimum at around 0.01% or 100 ppm sulphite concentration and for level higher than this there is a recovery in the swelling volume to a value approaching that of samples containing no sulphite. The reduction in swollen volume is accompanied by an increase in the amount of material found in the supernatant. This can be seen both from the measurement of total carbohydrates and total solids (Fig 2). If all the solid material was evenly distributed throughout the centrifuge tube then a solids content of about 9 mg/ml would be expected. This is what is observed at 0.01% sulphite addition. There are also changes in the clarity of the supernatant as evidenced by the change in absorbance. At higher levels of sulphite the appeared much clearer. (Fig 3).
Figure 1. The effect of sodium sulphite addition on the swelling volume of cassava starch. The error bars are standard errors of three replicates. The inset graph is for values of swelling volume for sulphite concentration between 0 and 0.001%
Figure 2. The effect of sodium sulphite addition on the solubility of cassava starch. The solubility is the total weight of solids of carbohydrate found in the supernatant expressed as a a percentage of the dry weight of starch in the whole system.
Figure 3. The effect of sodium sulphite addition on the clarity of cassava starch. The absorbance was measured at 330nm with distilled water as the reference sample. The inset graph is for values of absorbance for sodium sulphite concentrations between 0 and 0.01%
Effect of propyl gallate
In contrast, the addition of gallate at these low concentrations did not influence the swelling volume levels (Fig 4.). However, there is some evidence that even though gallate addition did not effect the swelling volume it did reduce the amount of material released into the supernatant (Fig 5). Since the amylose content of cassava is approximately 20%. It therefore appears that only this component was released into the supernatant during the pasting process when gallate was incorporated whereas in its absence some amylopectin may have been leached out. In order to be certain, it is necessary to characterize the supernatant polysaccharides in terms of its amylose-amylopectin ratio.
Figure 4. The effect of propyl gallate addition on the swelling volume of cassava starch.
Figure 5. The effect of proply gallate addition on the solubility of cassava starch. The solubility is defined in the legend for Figure 2 and is based on measurement of total carbohydrate content of the supernatant.
Effect of gallate and sulphite
When gallate was added to systems containing 0.01% sulphite, the pronounced reduction in swelling volume that was observed in the case of 0.01% sulphite alone, was prevented.
Figure 6. The effect of propyl gallate addition on the swelling volume of cassava starch containing 0.01% sodium sulphite.
The levels of gallate required to do this were only around 0.02% (Fig 6). The increase in swelling volume was accompanied by a decrease in the amount of carbohydrate released into the supernatant (Fig. 7).
Figure 7. The effect of propyl gallate addition on the solubility of cassava starch containing 0.01% sodium sulphite
Effect of the antioxidants on pH
It should be borne in mind that unbuffered solutions were used during tha pasting process. There were slight pH changes as a result of both the addition of these antioxidants and gelatinization. The mean pH value for the experiments was approximately 7.5 and nay variation amounted to no more than 1 pH unit change. It was also observed that in the case of sulphite alone, there was a reduction due to pasting. When gallate alone was added or in conjunction with sulphite the pH remained independent of the gallate concentration and the pH did not change after heating.
The results recorded demonstrate a number of features that require further consideration:
- There is a pronounced ,minimum in the swelling volume of cassava at around 0.01% (0.79 mmol/l) sulphite concentration. At levels above this the swelling volume recovers to a value approaching that of samples containing no sulphite.
- The addition of propyl gallate at 0.02% eliminated the pronounced reduction in swelling volume caused by 0.01% sulphite.
- Propyl gallate at 0.02% had no effect upon the swelling volume of cassava starch but did marginally reduce the amount of polysaccharides released into the supernatant.
More generally these points suggest that an explanation is required for the substantial leaching of the swollen granules promoted by sulphite and the mechanism of counteraction by antioxidants such as propyl gallate.
We shall first consider the possible role of sulphite. There is good evidence in the literature  that sulphite reacts stoichiometrically with oxygen (2 moles of sulphite with 1 mole of oxygen) resulting in the removal of the oxygen but leading to the formation of sulphite free radicals and the superoxide ion (.O2). More precisely this appears to involve an initiating stage probably catalyzed by trace transition metals in which molecular oxygen is transmitted to the sulphoxide ion:
M2↑ + O2→M3 + .O2
The scatter in the results, at very low levels of sulphite (Fig. 1) may well find an explanation in small variations in the levels of metal ions present. The process of propagation continues with the formation of sulphite free radicals.
SO32 + . O2 + 3H. → HSO3.+ 2.OH
SO32 + . O + 2H → HSO3. + H2O
HSO3. + O2 → SO3 + . O2+ H .
Finally, free radically may interact and terminate the propagation chain. This will be particularly evident where the reactants providing the source of free radicals are exhausted.
HSO3. + . OH → SO3 + H2O
2HSO3.→ SO3 + SO3 2 + 2H.
SO3 + H2O → SO42 + 2H
The oxygen concentration in water at 900C is 0.2 mmol/l . It would therefore seem that with the level of sulphite in the reaction mixture which produced the minimum in swelling volume (0.79 mmol)there will be an excess which will ensure the complete removal of molecular oxygen but with the accompanying production of superoxide ions and sulphite free radicals.
The possible effects of such free radicals upon the starch polysaccharides may be judged from the result of other workers. For example. Wellington  has demonstrated the important contribution of duperoxide ions in depolymerization processes. The specific activity of the sulphite free radical is less certain; it may, for example have an intrinsic activity of its own, or alternatively. It may provide by regeneration within the propagation sequence further supplies of superoxide ions.
With such a background, the effect of antioxidants may be interpreted in terms of their ability to inhibit free radical propagation processes. In line with the above reaction sequences Devand Jain observed that propyl gallate inhibits the oxidation of sulphite. It is therefore fully in accord with the second and third observations at the beginning of the discussion that propyl gallate may function either by directly inhibiting marginally significant ORD reactions, or secondarily by blocking the formation of superoxide and sulphite free radicals.
In our view this work has some interesting implications not only in terms of potential industrial applications in tarch technology but alos in assisting a development of our understanding of the changes taking place when the starch granule is heated. Both sulphite and gallate are allowed foor additives and thus os possible that they could be utilized in processes such as extrusion where starch degradation is recognized as being important . the rheology of dilute starch pastes depends on both the volume occupied by the swollen granule and the material leached out from the granule . It may be possible to control these two factors by appropriate combinations of antioxidants and we are currently investigating the rheology of such systems. It was reported in an early paper  that sulphite addition can reduce the viscosity of starch pastes although the levels of sulphite used were much higher. It is not yet clear whether the sulphite effect is specific to cassava starch or whether similar results will be found with other starches. The susceptibility of the granule to sulphite induced disintegration may provide information about the structure of starch. For example, the degree of entanglement between amylopectin molecules in what is essentially a swollen gel. Finally, although we have interpreted these results in terms of limited degradation of carbohydrate polymers through ORD reactions, as has been discussed above, the sulphite can promote lipid oxidation and this may inhibit the formation of amylose-lipid complexes which in the absence of sulphite have a stabilizing role on the thermal degradation of starch.
The effect of two oxidants, sodium sulphite and n-propyl gallate on the swelling-solubility characteristics of cassava starch was studied. It was found that the addition of sulphite dramatically reduced the swelling volume of cassava starch and promoted a near total solubilization of the polysaccharides into the continuous macro molecular phase. The addition of propyl gallate was effective in overcoming this effect it is postulated that oxidative reductive depolymerisation (ORD) reactions where responsible for causing the above behavior. These observations have significant implications for the industrial applications of starch. Studies on the Rheological properties of starch systems and on starches of other botanical origins are currently under way.