Studies on Dioscorea rotundata Starch Properties

S.N. Moorthy and S.G. Nair, Trivandrum (India)

Starch from six clonal selections of Dioscorea rotundata was isolated by standard procedure. The yield was between 20-24%. Various properties of these starches were compared. Granule size, 2% viscosity, peak viscosity, clarity, sol stability, total and soluble amylase contents were studied and only small variations were observed in the properties except paste viscosities among the clonal selections. The phosphorus content of Dioscorea starch was found to be three times as much as cassava starch, but low compared to potato starch. The higher gel strength of D, rotundata starch paste compared to cassava gel may be attributed to the phosphate linkages among the starch molecules in the granules.

Untersuchungen uber die Eigenschaften der Starke von Dioscorea rotundata. Die Starke von sechs klonalen Auswahlen von Dioscorea rotundata wurde mittels eines Standard-Verfahrens isoliert. Die Ausbeute betrug 20-24%. Verschiedene Eigenschaften der Starken wurden verglichen. KorngroBe, Viskositat 2% iger Suspensionen, Spitzenviskositat, Kalrheit. Solstabilitat, gesamte und losliche Amylose wurden untersucht und nur geringe Abweichungen in den Eigenschaftern beobachtet, abgesehen von den Kleisterviskositaten unter den klonalen Auswahlen. Der Phosphorgehalt von Dioscorea-Starke war dreimal so hoch wie der von Cassavastarke, aber niedrig im Vergleich zu Kartoffelstarke. Die hohere Gelstarke von D. rotundata-Starkekleister im Vergleich zu Cassava-Gelen durfte den Phosphatbindungen zwischen den Starkemolekulen in den Kornern zuzuschreiben sein.


Dioscorea rotundata is an important edible root crop grown widely in Africa and forms a prestigious component of the West African diet. Its taste is regarded as superior to that of other root crops [1]. However, there is little or no cultivation of D. rotunda in Asia [2]. Recently this crop has been introduced into India by the Central Tuber Crops Research Institute, Trivandrum, and has been found to adapt and yield well in different areas of the state of Kerala and promises to be a potential crop. Two selections, viz. Sree Subhra and Sree Priya, have recently been released by the Institute and they have been found to give a tuber yield of 35 to 40 t/ha and have acceptable quality.

The major component of the tubers is starch which is found to be over 75% of the dry matter content. There have been some studies on the D. rotundata starch, but systematic investigation on the properties in relation to varietial differences is lacking. The shape and size of the starch granules have been examined by some workers [3, 4]. Rasper found that the starch of most of the varieties he studied, possessed oval shape with an average length of 52 ?. The average granule size reported by Rodriguez- Soza and Parsi Ros is 33 ?m [4]. The amylase content was found to be around 21% [3]. The paste viscosity studies using Brabender amylograph have been carried out in detail by Rasper [3] and Rodriguez-Soza and Parsi Ros [4] and Rodriguez-Soza et al. [5]. They found that the starch has good viscosity stability and gel strength and also found that swelling power increased with temperature, reaching very high values at 95°C. There is possibility of wide variation in the properties of starch of different clones of D. rotundata. This paper deals with the study of various properties of starch of six clonal selections of D. rotundata. The results can throw various applications in addition to other tuber crop starches.

Material and Methods

The six clonal selections used for the study include two released varieties, viz. Sree Subhra (I-146), Sree Priya (U-195 (2)); two promising selections. (T3-(1) and 212), and two darf clones, 183 and 184. The crop was cultivated according to standard practices and harvested at 10 months stage.

Starch was extracted according to standard procedure [6]. The tubers were washed free of mud and peeled to remove the thin outer skin. The washed tubers were sliced and 1 kg of the material was crushed in a Waring Blender at room temperature using a large excess of water. The starch was allowed to settle and the supernatant decanted off. The starch cake was powdered and dried at 40-60°C for 24 h. The starch was accurately weighed and transferred to polythene covers for further studies. Microscope observations on the starch granules were carried out according to MacMasters [7]. Reducing values were obtained by Schoch’s procedure [8] and expressed as Ferricyanide numbers. The total amylase content was determined by method of Sowbhagya and Bhattacharya [9] and soluble amylase using procedure by Shanty et al.[10]. For 2% solution viscosity data, ISI procedure was followed and the same solution was used for determining the clarity and sol stabililty. Clarity was based on the absorbance by the solution at 500nm compared to water = 0, while sol stability was the time taken, in hours, after which the starch gel starts settling.

Brabender viscographic pattern 5% and 6% starch pastes was obtained on a Brabender viscoamylograph (model 80102) using 350 cm g cartridge. The speed of the rotor was fixed at 75 rpm and the rate of heating and cooling was 1.5°C/min. The viscosity at 97°, the viscosity after 20 min. stirring at 97°, the viscosity after cooling to 25° and the peak viscosity were read out from the curves and expressed as Barbender Units (B.U.). The pasting temperature was also obtained from the curves Swelling volume of the starch was determined at 95°C by standard procedure [11]. Six concentrations, 0.5, 1.0, 1.5, 2.0, 3.0 and 5% w/v were used in order to understand the effect of concentration on swelling properties. The phosphorus content of the starch samples was determined by the vanadomolybdate method [12].

Table 1 : Yield, Granule Size, Reducing Value and Amylose Content of D. rotundata Starch.

  Yield (%) Granule size (μ) Reducing value (Ferricyanide No.) Total amylase content (%) Soluble amylase content (%)
Range Av. Size
T3-1 21-33 10 -60 30.5 0.3 23.8 14.8
I-146 23.04 12.5-55 32.1 0.5 24.4 15.9
U-195(2) 20.43 15-50 34.3 0.6 24.6 14.1
I-212 23.9 10-75 29.5 0.3 22.6 15.9
183 23.1 10-50 27.2 0.6 21.7 15.9
184 23.1 7.5-57.5 30.8 0.6 20.9 15.0

Results and Discussion

The percentage of starch that could be recovered from the peeled tubers is given in Table 1. It is seen that the recovery of starch is quite high for all the cultivates, being above 20%. The highest value was for I-212 (23.9%). The values can be considered as good recovery, since the starch content of D. rotundata is around 23-25% on fresh weight basis, and, hence, there is little loss during extraction. The high recovery on overnight settling indicates that mucilaginous substances present do not interfere in starch settling. The results point out that the starch of D. rotundata can be easily and profitably extracted for various purposes. This can also give an impetus to the cultivation of the crop on a large scale.

The shape of the starch granules of the different cultivars was the same, viz. round to oval. Most of the granules were intact and single. This is similar to that reported by Rasper [3], who found oval granules in most of the varieties he studied. The size of the granule showed at wide variation, in the range 7.5-75?m (Table 1). There was no significant variation between the different samples. The average granule size varied from 27-34 ?m, with highest value being recorded for cultivar U-195 (2) and lowest for 183. These values are in the same range as reported earlier by other workers [3, 4]. Rasper found the size range to lie between 10-70 ?m, with a length 40-52 ?m and width 23-27 ?m [3]. Rodriguez-Soza and Parsi Ros obtained a value range 13-52 ?m with an average value of 33 ?m. The starch granules are thus smaller in size compared to D. alata and much bigger than D. esculenta starch granules [13]. The reducing values of the starches were found to be below 1.0 (Table 1) for all the cultivars indicating that the average molecular weight is high and similar to other tuber crop starches.

The total amylase content of the different cultivars was found to vary between 20.9 to 24.6%, with highest value for U-195 (2) (Table 1). The results are in conformity with the earlier reported values, 23-25% by Rasper [3] and 21-23% by Kay [14]. The amylase content is slightly higher than in cassava and D. esculenta starches and similar to D. alata starch. In spite of the higher amylase content, the starch does not behave as high amylase, being in the neighbourhood of 15% of the total starch. The soluble amylase content in D. rotundata starch has not been reported earlier.

The 2% viscosity of starch of the different cultivars showed variations between 37 and 46 s, highest being for U-195 (2) and lowest for 183 (Table 2). The Redwood viscosity values compare well with cassava starch, which usually has a value of 45-50 s. ISI has fixed a minimum value or 44 s for cassava starch for use as sizing agent in textile industries, and the starch of Dioscorea has viscosity very near to that of cassava.

The paste viscosities at 5% and 6% concentrations are given in Table 2. Notable variation in peak viscosity was observed between different cultivars with values ranging from 325 B.U. for starch of 184, to 550 B.U. for starch of U-195 (2) at 5% concentration. At 6% concentration, the values rose to 650 to 920 B.U. The data on viscosity values at 97°C, show that the complete gelatinization occurs in most cases after 97°. This indicates that the associative forces in the starch are strong and require prolonged heating to be disrupted before water enters. The viscosity after holding period is only a few units less than peak viscosity reflecting the strength of the starch paste. The gel strength is observed for all the cultivars and has been highlighted by earlier workers [3, 4]. The high gel strength is desirable in many food applications, since an unstable gel leads to a cohesive texture. Such high viscosity stability is not observed for cassava and sweet potato starches. The strong associative forces in the starch granules not only restrict the easy entry of water, but hold the swollen starch molecules together preventing breakdown.

Table 2 : Viscosity Properties and Pasting Temperature of D. rotundata Starch.

  2% Viscosity (s) Paste viscosity (B.U.) Pasting temperature (°C)
1) PV (5%) PV (6%) 2) V97 (5%) V97 (6%) 3) VH (5%) VH (6%) 4) Vc (5%) Vc (6%)
T3-1 37.5 470 730 450 720 480 770 620 1000 83-97
I-146 42.0 435 700 420 660 435 740 500 1000 81-97
U-195 (2) 46.0 550 920 570 900 550 1000 750 1000 80-97
I-212 45.0 500 770 490 760 500 800 770 1000 79-97
183 37.0 370 680 360 650 370 610 480 720 83-97
184 38.5 325 680 320 630 350 675 450 800 82-97

Table 3 : Swelling Volume, Clarity and Sol Stability of D. rotundata Starch

  Swelling Volume(ml/100ml) Clarity (Absorbance) Sol stability (h) P content (%)
  0.5% 1.0% 1.5% 2% 3% 5%      
T3-1 10.8 22.3 35.4 43.8 56.0 67.0 0.13 36 0.012
I-146 13.3 24.6 36.3 44.0 59.6 64.0 0.15 24 0.014
U-195(2) 12.5 25.0 36.6 48.6 59.2 66.0 0.16 24 0.013
I-212 12.9 23.8 35.0 45.0 59.6 67.0 0.14 36 0.011
183 8.2 15.0 21.9 30.8 45.0 62.5 0.10 24 0.015
184 7.5 15.0 22.5 30.7 43.5 62.5 0.12 24 0.015

The pasting temperature as given in Table 2 show only minor variation in the initiation point. However, I-212 showed a slightly earlier rise in viscosity, but it is not significant enough to infer that the starch of this cultivar has weaker associative forces.

The cooling curves exhibit only a slow and steady rise and no sudden rise characteristic of high amylase starches. This shows that the starch has only low retrogradation tendency similar to other tuber crop starches. This is, however, in contrast to the earlier report by Rodriguez-Soza who found a notable retrogradation tendency in Dioscorea starch. Rodriguez-Soza et al. found that the viscosity patten of D. rotundata starch is also pH dependent. The peak viscosity increased when pH was increased from 3 to 5, decreased till pH=7.0. The retrogratdation tendency was highest pH 5.5 and lowest pH 3[5].

The swelling volume of the starch of different cultivars as determined at 95°C for various concentrations is presented in Table 3. The swelling volume goes on increasing with increasing concentration until it tapers off above 2% concentration. At higher concentrations there is not enough water for the granule to imbide and swell. The largest swelling volume was obtained for I-146, at 0.5% concentration but at the higher concentrations, U-195(2) and I-146 possess higher swelling volumes. This is in line with the higher peak viscosity obtained for these starches. Lower swelling volumes were observed for starch of 183 and 184, which have lower peak viscosity. These starches may be possessing stronger associative forces. The values reported earlier [4] for the swelling power of D. rotundata starch at temperatures above 85°C were much higher with a correspondingly higher solubilitys. However, we could not obtain such high swelling volumes or solubility even at 95°C.

The clarity of the starch paste lies between those of cassava and potato starches. There is only minor variation between the different cultivars (Table 3). The clarity of the starch is similar to other tuber starches and unlike high-amylose starches.

The sol stability was low, but within acceptable ranges. Only very little difference was noticed among the different samples, the values being between 24-36 h. The retrogradation tendency observed in case of high amylase starches is not noticed in D. rotundata starch.

In view of the finding that the starch exhibits high gel strength, it was attempted to correlate the phosphorus content of D. rotundata starch with the gel strength. As seen in Table 3, the P content of this starch was three to four times that of cassava starch, but much lower than potato starch. It is already reported that the high gel strength and peak viscosity of potato starch is due to phosphate crosslinkages which are more difficult to break [15]. Probably, Dioscorea rotundata starch has also such phosphate crosslinkages, which render its properties like high gel strength, and viscosity stability different from those of cassava starch. However, this needs further confirmation.


  1. Horton, D., J. Lynam and H. Knipscheer: Proc. Sixth Symp. Intern. Soc. Trop. Root Crops, CIP, Lima, Peru, 1984, p.21.
  2. Onwueme, I.C.: Tropical Root and Tuber Crops, John Wiley and Sons, New York, p.234.
  3. Rasper, V.: Proc. First Intern. Symp. Trop Root Crops, Univ. West Indies, Trinidad, 2, 1967, p.48.
  4. Rodriguez-Soza, E.J., and O. Parsi-Ros: J. Agr. Univ. Puerto Rico 66 (1982), 27.
  5. Rodriguez-Soza, E.J., and O. Parsi-Ros, and M.A. Gonzlez: J. Agr. Univ. Puerto Rico 65 (1981), 154.
  6. Badehuizen, N.P.: In: Methods in Carbohydrate Chemistry, Vol.4, Ed. R.L. Whistler, Acad. Press, New York 1964, p.14.
  7. MacMasters, M.M.: In: Methods in Carbohydrate Chemistry, Vol.4. Ed. R.l. Whistler, Acad. Press, New York 1964, p.233
  8. Schoch, T.J.: In: Methods in Carbohydrate Chemistry, Vol.4. Ed. R.l. Whistler, Acad. Press, New York 1964, p. 64.
  9. Sowbhagya, C.M., and K.R. Bhattacharya: Starch/Starke 23 (1971), 53.
  10. Shanty, A.P., K. R. Bhattacharaya, and C.M. Sowbhagya: Starch/Starke 32 (1980), 409.
  11. Sair, L.: In: Methods in Carbohydrate Chemistry, Vol 4. Ed. R.L. Whistler, Acad. Press, New York 1964, p.284.
  12. Jackson, M.L.: Soil Chemical Analysis, Prentice Hall India Ltd., New Delhi 1968, p.498.
  13. Gallant, D.J., H. Bewa, Q.H. Buy, B.Bouchat, O.szylit, and L. Sealy: Starch/Starke 34 (1982), 255.
  14. Kay, D.E.: Root Crops Digest, TPI, London 1973, p.235.
  15. Leach, H.W.: In: Starch – Chemistry and Technology R.L. Whistler and E.F. Paschall, Acad. Press, New York 1965, p.297.

Dr. S.N. Moorthy and S.G. Nair, Scientist S2 (Organic Chemistry) and Scientist S2 (Genetics), respectively, at the Central Tuber Crops Research Institute, Sreekaryam, Trivandrum- 695 017, India.