Previous PageTable Of ContentsNext Page

Evaluation of alkaline extraction for obtaining starch and protein from Australian lentil cultivars, Matilda and Digger

H.C. Lee1, A.K. Htoon2 and J.L. Paterson1

1Food Science and Technology, School of Chemical Engineering and Industrial Chemistry, The University of New South Wales, Sydney NSW 2052, Australia
Food Science Australia (A Joint Venture of CSIRO and The Victorian Government), North Ryde, NSW 1670, Australia


Legumes are an important source of food for many people in various parts of the world. They are an excellent source of carbohydrate and provide an inexpensive source of protein (Jood et al., 1998). Lentils are divided into two main types based on differences between the seed size and cotyledon colour. Green lentils, which are also commonly known as brown, yellow, Chilean, Continental or Macrosperma lentils, are seed that have a green to brown seed coat with yellow cotyledons. The second types of lentils, red lentils (also known as Microsperma or Persian lentils) have a pale grey to dark seed coat with red cotyledons. The green Matilda, Laird and Invincible and the red Aldinga, Callisto, Cobber, Digger and Kye are common lentil varieties grown in Australia. Although extraction of legume starches has been extensively studied in the food industry, new legume starches will add variety to the current market. This study determines the proximate composition of starch and protein in two Australian lentils, Matilda and Digger, and evaluates the starch and protein yield from the alkaline extraction of its flour.

Materials and methods

Samples - Two Australian lentils, Matilda (Green lentil) and Digger (Red Lentil) were obtained from The Lentil Company (TLC), Horsham (Victoria). Each sample was ground into fine flour at The University of New South Wales using a Fitz hammer mill with a screen with aperture of about 0.79 mm.

Compositions - Moisture was determined by AACC Method 44-19 with modification. The samples were dried at 130C for 24 hours instead of the prescribed 135C. Estimation of crude protein (N x 6.25) was done using an automated LECO Nitrogen Analyser (LECO FD-428, LECO Corporation, St Joseph, MI, USA). Estimation of fat content was done by acid hydrolysis method: AOAC Official Method 922.06. The sugars and starch content were analysed using a method modified from Wills et al. (1980).

Extraction and Evaluations - The extraction of starch and protein from flour was carried out using a modified in-house extraction method adopted in Food Science Australia, CSIRO. Four temperatures (ambient 22C, 30C, 35C and 40C) and five different pH conditions (distilled water and pH 8.0, 8.5, 9.0 and 9.5) were used to determine the effect of alkaline extraction on starch and protein yield. Lentil flours and extracted starches were dispersed on double-stick adhesive tape mounted on SEM aluminium stubs, coated with thin layer of gold in EMITEX K 550X vacuum evaporator and examined with a FEI-QUANTA 200 Environmental Scanning Electron Microscope (ESEM) using a large field detector that operates in low vacuum mode. Starch damage was determined by the AACC method (76-31), using the starch damage assay kit purchased from Megazyme International, Ireland.

Results and discussion

Chemical Compositions

The composition of Digger and Matilda are given in Table 1. These values differed from values reported (Jood et al., 1998). The proteins content of the two Australian lentils were higher while the starch content was lower because cultivars and growing location differed from Jood’s samples. There was no significant difference between Matilda and Digger in terms of their starch and fat content. Matilda had a significantly higher protein content than Digger (P<0.05).

Table 1.Protein and starch content of lentil cultivars, Matilda and Digger (dry matter basis)


Moisture [%]a

Protein [%]b

Fat [%]b

Starch [%]c




10.65 0.12

32.62 0.27

2.76 0.11

45.00 5.52


11.77 0.21

30.33 0.21

2.86 0.07

44.77 2.16

a. Values are mean of SD of eight independent determinations
b. Values are mean of SD of five independent determinations
c. Values are mean of SD of six independent determinations

Evaluation of isolated lentils starch and protein by alkaline extraction

The extraction at pH 9.5, for all temperatures, produced the maximum starch yields (85-95%) for both varieties. The higher the extraction pH, the higher the % starch recovered. The effect of pH was more marked in Matilda than in Digger (Figure 1). The yields of Digger starch were not significantly different when compared within the four alkaline pHs across all 4 temperatures. At 95% confidence level, the recoveries from the four high pHs were not significantly different at any temperature (Figure 1B). In contrast, Matilda flour was more sensitive to pH and temperature. It showed gradual increase in % starch yield with increase in extraction temperatures and pH conditions (significant to 95% confidence level) (Figure 1A). Extraction temperatures had less effect on the starch yield in Digger than in Matilda. However, the increase in yield of Matilda starch with increase in extraction temperature was significant (P<0.05).



Figure 1. Starch yield of extraction vs temperature for: (A) green lentil, Matilda; (B) red lentil, Digger

Protein yield achieved from both Digger and Matilda flour was relatively low, ranging from about 43-60% of the analysed protein content for Digger and 48-63% for Matilda. In Digger, a significant increase in % protein yield was observed under alkali conditions as compared to distilled water; in all extraction temperatures. However, the yields of Digger proteins were not significantly different when compared within the four alkaline pHs across all four temperatures (P<0.05) (Table 1).

Table 1. Statistical summary for the effect of pH conditions on % protein yield for Digger





Distilled Water

43.78 0.75a

44.81 3.37a

49.26 1.11ab

48.79 0.15a

pH 8.0

54.33 4.14bcde

56.34 2.14bcde

57.87 4.24abcde

60.24 1.75bcde

pH 8.5

54.33 2.05 bcde

57.62 3.51 bcde

59.29 1.40 bcde

60.39 1.29 bcde

pH 9.0

56.72 1.32 bcde

58.01 1.24bcde

59.49 1.03 bcde

60.97 0.83 bcde

pH 9.5

59.10 2.53bcde

59.58 2.27bcde

60.34 3.55bcde

61.97 2.01bcde

Values are means of duplicate analyses SD
Superscripts: a-Distilled Water; b-pH 8.0; c-pH8.5; d-pH 9.0; e-pH 9.5
Means within a column followed by different superscripts are significantly different (P<0.05) at the compared pH conditions

Table 2. Statistical summary for the effect of pH conditions on % protein yield for Matilda





Distilled Water

48.49 0.65a

49.12 0.40a

50.89 0.06a

54.22 1.40a

pH 8.0

51.91 0.40bdc

53.49 0.02bc

60.34 0.93bcde

60.32 1.03bcd

pH 8.5

52.39 1.55bdc

52.70 0.69bc

59.02 0.24bc

60.39 0.11bc

pH 9.0

55.83 2.58bcde

56.60 1.49de

60.76 0.23bde

62.42 0.01bd

pH 9.5

60.31 0.54de

59.90 1.85de

60.66 0.35bde

63.36 0.32e

Values are means of duplicate analyses SD
Superscripts: a-Distilled Water; b-pH 8.0; c-pH8.5; d-pH 9.0; e-pH 9.5
Means within a column followed by different superscripts are significantly different (P<0.05) at the compared pH conditions

Matilda protein extraction differed from Digger. Across all four extraction temperatures, there was no significant difference for % protein yield at pH 8.0 and 8.5. However, most of the other % protein yields achieved were significantly different at different extraction pHs at the same extraction temperature (Table 2). Thus, extraction pH condition affects the protein extraction efficiency of Matilda more than Digger. The same trend was observed in both Digger and Matilda when evaluating the effect of temperature on protein extraction efficiency. Matilda was more sensitive to the effect of temperature than was Digger. Most of the % protein yield achieved for Digger was found to be not significantly different (P<0.05) when comparing the yield achieved within the various extraction temperatures at the same extraction pH. In the case of Matilda, significant differences in protein yield were observed. However, this effect was lesser at high pH condition (pH 9.5), as the yields achieved across three extraction temperatures (22οC, 30C and 35C) were not significantly different (P<0.05).

Figure 2. SEM micrograph of starch extracted using pH 9.5 at 40C at 2000x magnification for: (A) Digger and (B) Matilda. Deformed starch granules (d) and presence of cracks within granules (c)

The % nitrogen values obtained for all the alkaline extracted starches for both varieties were not detected when analysed using the LECO protein analyser. This denotes the absence of protein in the extracted starch up to a sensitivity range of 0.001%. In addition, images from SEM confirmed there was no visible protein among the starch granules (Figure 2). The starch granules achieved from the extraction process were clean. However, some residual materials, probably protein, were seen in starch extracted using distilled water. No residual materials were observed on the starch granules from the starch extracted at high pH.

In order to ensure good functional starch properties, high % starch damage should be avoided. Although high extraction pH (9.5) and temperature (40C) gave the best yield, it caused high % starch damage (Figure 3). All extraction conditions (high pH, high temperature) that resulted in starch damage of greater than 1.0% were eliminated. In addition, low extraction pHs (distilled water and pH 8.0) were eliminated even though % starch damage was low as these two conditions gave relatively low starch and protein yield. Taking all factors into account, pH 9.0 at 30C was chosen as an optimum extraction condition for Matilda and pH 8.5 at 35C was chosen for Digger.



Figure 3. Starch damage vs temperature for: (A) red lentil (Digger); and (B) green lentil (Matilda)


Lentil flour was extracted to produce clean starch and protein fractions with acceptable yield by alkaline extraction. The optimum condition for alkaline extraction of the starch from both varieties of Australian lentils was pH 9.0 at 30C for Matilda and pH 8.5 at 35C for Digger.


Jood, S., Bishnoi, S. and Sharma, A. (1998). Nahrung 42(2): 71-74.

Wills, R. B. H., Balmer, N. and Greenfield, H. (1980). Food Technology in Australia 32(4): 198-202, 204

Previous PageTop Of PageNext Page