Research Article Received: 2 March 2010
Revised: 13 May 2010
Accepted: 28 May 2010
Published online in Wiley Interscience: 9 July 2010
(www.interscience.wiley.com) DOI 10.1002/jsfa.4061
Sensory quality of marama/sorghum composite porridges ´ Eugenie Kayitesi,a,b Kwaku Gyebi Duodu,a Amanda Minnaara and Henriette L de Kocka∗ Abstract BACKGROUND: The edible seeds of marama beans are a good source of protein and fat and can potentially enhance the nutritional quality of sorghum products. Utilisation of marama flour in a composite porridge depends on sensory acceptance. Heating of marama beans before milling is used to inactivate trypsin inhibitors. Defatting of marama flour would improve shelf life. In this study, marama/sorghum composite porridges (10% solids) were prepared and subjected to descriptive sensory analysis, consumer acceptance testing, texture analysis, pasting and colour measurements. RESULTS: Composite porridges had less cooked sorghum aroma/flavour but more intense overall aroma/flavour and were less viscous and less firm than sorghum porridge. Composite porridges prepared with flour from heated marama beans had a roasted nut flavour and were darker in colour compared with porridges prepared with unheated marama beans, which had a strong boiled nut aroma/flavour. Composite porridges from full-fat marama flours tasted more buttery/creamy. A bitter aftertaste was perceived in porridges from defatted marama flours. The 100% sorghum porridge and the composite porridge with full-fat flour were more acceptable to consumers. CONCLUSION: Combining sorghum meal with full-fat marama bean flour has the potential to improve the sensory quality of sorghum porridge as a staple by providing acceptable sensory appeal. c 2010 Society of Chemical Industry Keywords: marama; sorghum; composite; flour; porridge; sensory quality
INTRODUCTION
2124
Marama bean (Tylosema esculentum (Burch) A. Schreib) is an underutilised wild legume growing in southern Africa. The marama bean is potentially a valuable crop for semi-arid countries1 and has long been identified as a food source.2 It is consumed by various rural communities, especially people of the Kalahari Desert. Marama beans are boiled as a green vegetable similar to green peas. The dry beans are roasted and eaten as snacks.3 The edible seeds are a good source of protein and fat and can potentially enhance the nutritional quality when incorporated into sorghum foods, a staple in many African countries. Sorghum has low protein content and is deficient in the essential amino acid lysine. Combining sorghum meal with marama flour could therefore improve the nutritional quality of the resultant porridge. Utilisation of marama flour in composite porridges will depend not only on nutritional quality but also on sensory acceptability of the porridge. Different processing procedures used in the production of marama flour may affect the sensory quality of food products incorporating the flour. Marama beans are rich in unsaturated fatty acids3 and thus the milled flour could also be susceptible to oxidation. Defatting of marama flour is done to improve its storage stability. However, removal of fat may lead to loss of its inherent sensory properties. This may affect the acceptability of its products such as porridge. Dry heating of marama beans prior to flour processing is aimed at deactivating protease inhibitors such as trypsin inhibitors that have
J Sci Food Agric 2010; 90: 2124–2132
a negative effect on protein digestion by humans.4 Heat treatment of marama beans leads to flavour and colour development, which makes them more palatable.3 The objective of this study was to determine the effects of dry heating marama beans, defatting and combining the resultant flours with sorghum meal on the sensory quality of porridges made from the flours.
MATERIALS AND METHODS Preparation of marama flours Marama flours were prepared using the procedure described by Maruatona5 with some modifications. Clean marama beans were dry heated (150 ◦ C for 20 min) in a forced convection, continuous tumble roaster (Roastech, Bloemfontein, South Africa). Heated and non-heated marama beans were dehulled separately with a DF cracker (WMC Metal Sheet Works, Tzaneen, South Africa). Dehulled marama beans were coarsely ground in a Waring laboratory blender (Lasec, Johannesburg, South Africa) and defatted by stirring in hexane (flour/hexane 1 : 3 w/v) for 2 h. This was repeated
∗
Correspondence to: Henriette L de Kock, Department of Food Science, University of Pretoria, Pretoria 0002, South Africa. E-mail:
[email protected]
a Department of Food Science, University of Pretoria, Pretoria 0002, South Africa b Department of Food Science, Kigali Institute of Science and Technology, Kigali 3900, Rwanda
www.soci.org
c 2010 Society of Chemical Industry
Sensory quality of marama/sorghum composite porridges three times, decanting off the hexane each time. The defatted flour was left in a fume cupboard for 24 h to evaporate off excess hexane and then milled in a laboratory mill (IKA Werke, Staufen, Germany) to pass through a 1 mm mesh. Four marama flour samples, full-fat heated (FH), full-fat unheated (FUH), defatted heated (DH) and defatted unheated (DUH), were prepared, vacuum packed and stored at 4 ◦ C until combined with sorghum meal. Preparation of marama/sorghum composite flours Commercial sorghum meal (Monati Super Mabele, Nola Foods, Randfontein, South Africa) was purchased at a supermarket in Pretoria, South Africa. Full-fat and defatted marama flours from heated and unheated marama beans were combined with sorghum meal (marama flour/sorghum meal 30 : 70 w/w), vacuum packed and stored at 4 ◦ C until used to make porridges. A commercially available sorghum/soya composite porridge (Tsabotlhe, Foods Botswana, Gaborone, Botswana) was used as a standard. Pasting properties of marama/sorghum composite flours The pasting properties of flour suspensions containing 4 g of flour in 25 g of water were studied using a Rapid Visco Analyser (RVA) (Newport Scientific, Warriewood, Australia). Preparation of porridges Six soft porridges (10% solids) were prepared using (1) 100% sorghum meal (SOGP), (2) full-fat flour from unheated marama beans combined with 70% sorghum meal (FUMSP), (3) full-fat flour from heated marama beans combined with 70% sorghum meal (FHMSP), (4) defatted flour from unheated marama beans combined with 70% sorghum meal (DUMSP), (5) defatted flour from heated marama beans combined with 70% sorghum meal (DHMSP) and (6) Tsabotlhe (TBTP). A Botswana porridge-making process6 with some modifications was used for the preparation of soft porridges. The cooking process entailed first mixing 200 mL of cold water with 80 g of flour to make a slurry. The slurry was then gradually added, while stirring to avoid lump formation, to 600 mL of boiling water in a 2 L stainless steel saucepan. The porridge was simmered on low heat (on a hot plate) for 20 min, stirring every 5 min. Sensory panel Students from the University of Pretoria who were willing to consume sorghum porridge, did not suffer from any food allergies and had some experience of descriptive sensory evaluation were screened for sensory acuity. A panel of seven experienced judges was trained for 9 h following the generic descriptive analysis method described by Einstein.7 During the training, each panellist described the differences between the six porridge samples at least three times. Descriptive terms and scale anchors were developed, defined and agreed upon for evaluation (Table 1). Before the actual evaluation the panellists’ performance was checked at least three times. The FCM tool in Compusense five Release 4.8 (Compusense, Guelph, Canada) was used to facilitate training.
J Sci Food Agric 2010; 90: 2124–2132
laboratory with individual booths equipped with computers for direct data entry using Compusense five Release 4.8. Panellists evaluated all samples in triplicate for 3 days, with one session per day. Each panellist received six samples of porridge on a white tray, with six stainless steel teaspoons, a serviette and a plastic disposable cup filled with filtered tap water for rinsing the mouth before and between tasting the samples. To avoid fatigue, three samples were evaluated and after a 10 min break the other three samples were evaluated. The order of sample presentation was randomised over the panel. Red light in the tasting booths was used to mask the colour of the porridges in order to allow the panellists to concentrate on aroma, texture and flavour. The panel used 16 descriptive terms grouped under aroma, flavour, texture and aftertaste attributes. Aroma was evaluated using short sniffs immediately after removing the foil cover. Then a full spoon of the porridge was chewed in the mouth to evaluate flavour and texture. After swallowing, the panellists analysed aftertaste. A structured line scale with ten demarcated points was used to measure the intensity of each attribute for a given sample. The minimum value was 0, denoting not perceived, not coarse or not thick. The maximum point was 9, denoting strongly perceived, very coarse or very thick. Textural properties of porridges The textural properties (firmness and stickiness) of the porridges were determined using a TA-XT2 texture analyser (Stable Micro Systems, Godalming, UK). The following instrumental test parameters were used: mode was force in compression; option return to start; pre-test speed was 2 mm s−1 ; test speed was 2 mm s−1 ; post-test speed was 10 mm s−1 ; sample penetration distance was 10 mm; trigger type was auto-0.05 N; and a flat cylindrical perspex probe (20 mm diameter) was used. Porridge samples were prepared by gradually adding flour slurry (50 g of flour in 100 mL of cold water) to 400 mL of boiling water in a small stainless steel saucepan on an electric stove set on low heat, stirring continuously to avoid lump formation. The porridge was simmered for 15 min (stirring every 5 min) and then immediately filled into three sample tubes (50 mL volume and 30 mm diameter). The tubes were covered with aluminium foil and held at 50 ◦ C for 90 min. To determine the texture of the porridges, the aluminium foil was removed from the tube and the surface layer of the porridge was scraped off. The tube was then firmly secured centrally on the texture analyser stage. The test cycle was started immediately and the force–time curve was recorded. Two parameters were derived from the curve: porridge firmness, defined as the maximum force obtained as the probe penetrated the porridge, and porridge stickiness, defined as the maximum force recorded as the probe withdrew from the porridge. Colour measurements of porridges The colour of porridges prepared as described above for descriptive sensory analysis was measured using a Chroma Meter CR-400 (Konica Minolta Sensing, Osaka, Japan). The colour measurements were expressed in terms of lightness (L∗ ), red/green characteristics (a∗ ) and blue/yellow characteristics (b∗ ). Consumer sensory evaluation of porridges The six porridge samples were prepared and presented to consumers in the same way as for descriptive sensory analysis. South African consumers (n = 52), 30 males and 22 females, who ate sorghum porridge at least once a week and reported no
c 2010 Society of Chemical Industry
www.interscience.wiley.com/jsfa
2125
Sample evaluation Cooked porridge (40 g portions) was served in glass ramekins covered with aluminium foil and kept warm on a Salton warming tray (Salton House, Manchester, UK) at 50 ◦ C. The sensory evaluation of the porridge was conducted in a sensory evaluation
www.soci.org
www.soci.org
E Kayitesi et al.
Table 1. Terminology used by descriptive sensory panel to describe sensory attributes of porridges Attribute Aroma Overall aroma strength Cooked sorghum porridge aroma Boiled nut aroma Roasted nut aroma Flavour Overall flavour strength Cooked sorghum porridge flavour Boiled nut flavour Roasted nut flavour Buttery/creamy flavour Salty Bitter Texture Coarseness
Viscosity Oily/creamy Aftertaste Sorghum aftertaste
Definition
Reference
Bland = 0, strong = 9
The overall aroma intensity of the porridge Intensity of aroma associated with cooked sorghum porridge Intensity of aroma associated with boiled nuts Intensity of aroma associated with medium-roasted nuts and having fragrant character such as methyl pyrazine
Cooked sorghum porridge (Monati Super Mabele) rated 8 Boiled whole marama bean seeds (boiled for 45 min) rated 7 Roasted dehulled marama bean seeds (150 ◦ C for 20 min) rated 8
Not perceived = 0, strongly perceived = 9 Not perceived = 0, strongly perceived = 9 Not perceived = 0, strongly perceived = 9
Bland = 0, strong = 9
The overall aroma intensity of the porridge Intensity of flavour associated with cooked sorghum porridge
Cooked sorghum porridge (Monati Super Mabele) rated 8
Not perceived = 0, strongly perceived = 9
Intensity of flavour associated with boiled nuts
Boiled whole marama bean seeds (boiled for 45 min) rated 7 Roasted marama beans (150 ◦ C for 20 min) rated 8
Not perceived = 0, strongly perceived = 9 Not perceived = 0, strongly perceived = 9
Unsalted Clover butter rated 7
Not perceived = 0, strongly perceived = 9 Not perceived = 0, strongly perceived = 9 Not perceived = 0, strongly perceived = 9
Intensity of flavour associated with medium-roasted nuts and having fragrant character such as methyl pyrazine Intensity of flavour associated with butter Basic salt taste associated with sodium chloride (table salt) Basic bitter taste associated with caffeine or quinine Extent to which grittiness or graininess of porridge caused by small particles could be perceived in the mouth Force required to draw a liquid from a spoon over the tongue Intensity of mouthfeel associated with fatty or oily products such as dairy cream
Sour aftertaste
Intensity of aftertaste associated with cooked sorghum porridge perceived after swallowing the porridge Basic bitter taste associated with caffeine or quinine Basic sour/acidic taste associated with citric acid
Sweet aromatic nutty aftertaste
Sweet nutty aftertaste associated with roasted peanuts
Bitter aftertaste
Rating scale
food allergies were invited to participate by trained field workers. Each porridge sample was evaluated for overall acceptability on a nine-point hedonic scale.8 The nine structural levels ranged from 9 ‘like extremely’, through 5 ‘neither like nor dislike’, to 1 ‘dislike extremely’. Questions and scales were displayed on the computer screens using Compusense five Release 4.8. Consumers were also prompted to provide comments on the reasons why they liked or disliked the different porridges.
2126
Statistical analyses With the exception of the consumer acceptability test, which was not repeated, all analyses were conducted in triplicate. Analysis of variance (ANOVA) based on a 5% level of significance was used to test for the effect of sample on the measured parameters.
www.interscience.wiley.com/jsfa
0.35% sodium chloride in spring water (rated 9) 0.15% caffeine in spring water (rated 9)
Not coarse = 0, very coarse = 9 Thick cooked maize porridge (12% solids) rated 9
Not thick = 0, very thick =9 Not perceived = 1, strongly perceived = 9
Cooked sorghum porridge (Monati Super Mabele) rated 6
Not perceived = 1, strongly perceived = 9
0.15% caffeine in spring water rated 7
Not perceived = 0, strongly perceived = 9 Not perceived = 0, strongly perceived = 9 Not perceived = 0, strongly perceived = 9
0.08% citric acid in spring water rated 9 Roasted peanuts rated 7
The main effects of sample and panellist were included in the ANOVA of the sensory data. Statistica Version 8.0 (Statsoft, Tulsa, OK, USA) was used for data analysis. Principal component analysis (PCA) was used to evaluate the multivariate factors distinguishing the sensory profiles of the porridges. For consumer acceptability testing, box and whisker plots were used to demonstrate the distribution of ratings for the porridges.
RESULTS Referring to our work on nutritional quality evaluation, the fat and protein contents (w/w) of the porridges were as follows: SOGP, 11.7% protein and 1.4% fat; FUMSP, 18.8% protein and 12.5% fat;
c 2010 Society of Chemical Industry
J Sci Food Agric 2010; 90: 2124–2132
Sensory quality of marama/sorghum composite porridges
www.soci.org
Table 2. Effect of dry heating marama beans, defatting and combining resultant flour with sorghum meal on colour (L∗ , a∗ and b∗ values)a of porridges Type of porridgeb SOGP FHMSP FUMSP DHMSP DUMSP
L∗
a∗
b∗
58.8a (0.5) 62.1b (0.4) 67.4d (0.7) 63.2c (0.4) 68.5e (0.8)
2.8a (0.3) 4.5c (0.4) 2.9a (0.2) 3.9b (0.1) 2.2a (0.2)
5.1a (0.2) 12.5d (0.5) 8.4c (0.5) 13.0d (0.2) 7.1b (0.4)
Mean values with different letters within a column differ significantly (P < 0.05); standard deviations are given in parentheses. a L∗ , lightness (0 = black, 100 = white); +a∗ , red; −a∗ , green; +b∗ , yellow; −b∗ , blue. b SOGP, sorghum porridge; FHMSP, full-fat heated marama/sorghum porridge; FUMSP, full-fat unheated marama/sorghum porridge; DHMSP, defatted heated marama/sorghum porridge; DUMSP, defatted unheated marama/sorghum porridge.
J Sci Food Agric 2010; 90: 2124–2132
DISCUSSION Consumer acceptability of the porridges was much dependent on the differences in the sensory attributes of the porridges in terms of flavour, aroma, colour and texture as quantified by the descriptive sensory and instrumental analyses. The 100% sorghum porridge (SOGP) and the marama/sorghum porridge made with full-fat flour from heated beans (FHMSP) were the most acceptable. This may be because the sensory properties of SOGP were more familiar to the consumers, as it is at present the product mostly consumed. FHMSP was probably
c 2010 Society of Chemical Industry
www.interscience.wiley.com/jsfa
2127
FHMSP, 18.8% protein and 12.7% fat; DUMSP, 22.9% protein and 2.9% fat; DHMSP, 22.6% protein and 2.7% fat. The composite porridges from unheated marama beans (DUMSP and FUMSP) were significantly lighter (as indicated by higher L∗ values) than those from heated marama beans (DHMSP and FHMSP), but generally all marama composite porridges were lighter than the sorghum porridge (Table 2). Table 2 also shows that DHMSP and FHMSP had relatively higher a∗ and b∗ values and therefore more colour than DUMSP and FUMSP. Combining marama flour with sorghum meal affected the aroma and flavour of the porridge (Fig. 1). The sorghum porridge (SOGP) had a more intense cooked sorghum aroma and flavour compared with the composite porridges. DHMSP and FHMSP had a significantly more intense roasted nut aroma and flavour than DUMSP, FUMSP, TBTP and SOGP. The aroma and flavour of DUMSP and FUMSP were described as boiled nut. The latter porridges were closely comparable to Tsabotlhe (soya/sorghum porridge) in terms of overall aroma and flavour strength. Introduction of legume flours (marama or soya flour) significantly increased buttery/creamy flavour and oily mouthfeel in the composite porridges. TBTP was perceived as more salty than the rest of the porridge samples. Bitterness was perceived in the composite porridges with defatted marama flour (DUMSP and DHMSP). A sweeter aromatic nutty aftertaste was perceived in the marama/sorghum composite porridges than in the sorghum porridge. The composite porridges had lower viscosity than the sorghum porridge (Table 3). TBTP had the lowest viscosity score. FHMSP, FUMSP and TBTP were coarser than the other porridges. The lower viscosity of the composite porridges compared with the sorghum porridge is also shown by the results for pasting properties (Table 4). Pasting is a process that follows gelatinisation, involving swelling of starch granules, leaching and alignment of granule components and disruption of starch granules.9 The pasting profiles of the marama/sorghum composite porridges displayed the same general pattern, with small variations. Peak viscosity, trough, breakdown, setback and final viscosity of the marama/sorghum composite porridges were significantly lower (P < 0.05) than those of the sorghum porridge (Table 4). Figure 2 shows that the composite porridges did not differ significantly in firmness and stickiness among themselves. However, the sorghum
porridge was firmer and slightly stickier than the composite porridges. PCA was used to summarise the variation in the descriptive sensory attributes of the porridges. Figure 3A shows the projection of the scores of the porridges and Fig. 3B illustrates the loading projections of the sensory attributes. The first two principal components described 76% of the total variation in sensory attributes of the porridges. The first principal component (PCA 1) explained 52% of the total variation. The sorghum porridge on the right side of the plot was differentiated from the marama/sorghum and soya/sorghum composite porridges. Cooked sorghum aroma, flavour and aftertaste were strongly perceived in the sorghum porridge; furthermore, the sorghum porridge was more viscous than the composite porridges. Porridges on the left side of the plot included the full-fat and defatted marama/sorghum porridges from heated and unheated marama beans, which were characterised by the attributes roasted nut and boiled nut aroma and flavour, buttery/creamy flavour, oily mouthfeel and sweet nutty aromatic aftertaste. The second principal component (PCA 2) added 24% to the explanation of the variation and separated DHMSP and FHMSP, with more roasted nut aroma and flavour, overall aroma and flavour strength, at the top of the plot from DUMSP and FUMSP, which were characterised by the attributes boiled nut aroma and flavour. The scores for the standard soya/sorghum porridge (TBTP) on PCA 1 and PCA 2 were close to zero, indicating that these two principal components did not adequately differentiate this porridge from the others. The third principal component (PCA 3) explained an additional 13% (making a total of 89%) of the variation and differentiated the marama/sorghum porridges from defatted flour, which were slightly more bitter compared with the other porridges (Figs 3C and 3D). PCA 3 also showed that TBTP was more salty. Mean ratings illustrating the effects of dry heating marama beans, defatting and combining the resultant flours with sorghum meal on consumer acceptability of the porridges are presented in Fig. 4. The distribution along the line bar on the graph explained the agreement among consumers: the shorter the score distribution, e.g. as shown for SOGP, the more the agreement among consumers. The widest variation in opinions on the acceptability of a porridge is shown for TBTP. This indicated that some consumers appreciated the sensory properties of this porridge while others disliked it very much, as illustrated by comments such as ‘smooth and soft’ and ‘salty and watery’ respectively. SOGP was as acceptable to consumers as FHSMP. DUMSP and DHMSP, while TBTP had the lowest mean score of all the porridges. Consumers noted DUMSP and DHMSP as ‘bitter’ and TBTP as ‘salty and watery’ as some of the reasons for giving low ratings.
www.soci.org
Bitter
c bc
a ab
a a
E Kayitesi et al.
ab ab b
ab
Salty
c
a
b b
bc
bc
Buttery/creamy flavour
c
a c a a
Roasted nut flavour
c b
a
b
d b
Boiled nut flavour
e
c a a a
Cooked sorghum porridge flavour
a a b
c b b b
a ab
Overall flavour strength
ab c ab Roasted nut aroma
ab
c b
a b
cd c
e
Boiled nut aroma
d a a a
Cooked sorghum porridge aroma
a a a
b ab b
Overall aroma strength a 0
1
2
3
4
5
c bc bc 6
Intensity scale Figure 1. Aroma and flavour sensory attributes (0 = not perceived, 9 = strongly perceived) of porridges: SOGP, sorghum porridge; TBTP, Tsabotlhe FUMSP, full-fat unheated marama/sorghum porridge; FHMSP, full-fat heated (sorghum/soya porridge) commercial product from Botswana; marama/sorghum porridge; DUMSP, defatted unheated marama/sorghum porridge; DHMSP, defatted heated marama/sorghum porridge. For each attribute, mean values with different letters differ significantly (P < 0.05).
2128 www.interscience.wiley.com/jsfa
c 2010 Society of Chemical Industry
J Sci Food Agric 2010; 90: 2124–2132
Sensory quality of marama/sorghum composite porridges
www.soci.org
Table 3. Effect of dry heating of marama beans, defatting and combining resultant flour with sorghum meal on texture/mouthfeel and aftertaste attributes of porridgesa Attribute Porridge viscosity Coarseness Oily mouthfeel Sorghum flavour aftertaste Sweet aromatic nutty aftertaste Sour aftertaste Bitter aftertaste
SOGP
TBTP
FUMSP
FHMSP
DUMSP
DHMSP
6.0c (1.0) 2.6a (0.9) 0.9a (0.8) 4.5c (1.2) 0.9a (0.8) 0.7a (0.7) 0.8a (0.7)
3.3a (0.7) 3.9b (1.3) 2.9bc (1.0) 3.0b (1.5) 2.8bc (1.0) 0.7a (0.7) 0.8a (0.6)
4.3b (0.8) 5.3c (0.7) 3.5c (1.3) 2.4ab (1.3) 3.9d (1.1) 0.7a (0.8) 0.8a (0.7)
4.4b (0.7) 4.2b (0.8) 3.5c (1.6) 2.4a (1.8) 3.7d (1.0) 0.7a (0.7) 0.9a (0.9)
4.1b (1.2) 2.5a (0.9) 2.6b (0.9) 2.5ab (1.2) 2.7b (1.0) 0.7a (0.7) 1.0ab (0.8)
3.9b (0.9) 2.2a (1.0) 3.0bc (1.3) 2.3a (1.2) 3.3cd (1.3) 0.7a (0.7) 1.2b (0.9)
Mean values with different letters within a row differ significantly (P < 0.05); standard deviations are given in parentheses. a SOGP, sorghum porridge; TBTP, Tsabotlhe (sorghum/soya porridge) commercial product from Botswana; FUMSP, full-fat unheated marama/sorghum porridge; FHMSP, full-fat heated marama/sorghum porridge; DUMSP, defatted unheated marama/sorghum porridge; DHMSP, defatted heated marama/sorghum porridge.
Table 4. Pasting properties (Rapid Visco Units) of porridges Type of porridgea
Peak viscosity
Trough
Breakdown
Setback
Final viscosity
SOGP FUMSP FHMSP DUMSP DHMSP
245.6e (4.5) 90.2b (0.6) 72.8a (3.6) 109.3c (1.8) 122.4d (10.0)
218.2c (5.1) 73.8b (0.5) 63.0a (2.1) 76.8b (0.9) 73.0b (4.9)
27.4b (9.2) 16.4ab (0.3) 9.7a (1.5) 32.5b (0.9) 49.4c (5.2)
290.9c (21.4) 109.8ab (3.2) 106.6ab (4.8) 92.0a (0.9) 135.9b (7.5)
509.1c (16.5) 183.6ab (3.7) 169.6a (6.9) 168.7a (1.9) 208.9b (12.4)
Mean values with different letters within a column differ significantly (P < 0.05); standard deviations are given in parentheses. a SOGP, sorghum porridge; FUMSP, full-fat unheated marama/sorghum porridge; FHMSP, full-fat heated marama/sorghum porridge; DUMSP, defatted unheated marama/sorghum porridge; DHMSP, defatted heated marama/sorghum porridge.
Force (N) 0.8
SOGP
d1f
DUMSP 0.6
Firmness (+ve force)
FUMSP
0.4
FHMSP DHMSP
0.2
-0.0 0
-0.2
-0.4
2
4
6
8
10
12 Time (s)
Stickiness (–ve force) d2f 2
-0.6
J Sci Food Agric 2010; 90: 2124–2132
c 2010 Society of Chemical Industry
www.interscience.wiley.com/jsfa
2129
Figure 2. Firmness (maximum force obtained as probe penetrated porridge) and stickiness (maximum force recorded as probe withdrew from porridge) of porridges: SOGP, sorghum porridge, FUMSP, full-fat unheated marama/sorghum porridge; FHMSP, full-fat heated marama/sorghum porridge; DUMSP, defatted unheated marama/sorghum porridge; DHMSP, defatted heated marama/sorghum porridge.
www.soci.org more acceptable because of its rich, buttery and creamy flavours. Fat is a carrier of many flavour compounds and contributes its own natural richness and creaminess.10 Roasted nut aroma and flavour and sweet aromatic nutty aftertaste characterised FHMSP and DHMSP. This is attributed to desirable flavours obtained from Maillard reaction products such as pyrazines produced during dry heating.11 Bitter taste and aftertaste were present, although at low levels, in DUMSP and DHMSP. A decrease in fat content induces an increase in flavour release by high-molecular-weight compounds.12 Removal of fat probably unmasked the bitter taste perceived in these porridges. Bitterness might also be associated with the presence of phenolic compounds that cause bitterness in legumes, as reported by Drewnowski and Gomez-Carneros.13 Bitterness was more prominent in porridges from the heated samples. Bitterness due to excessive heat treatment of marama beans was reported by Mmonatau.3 Bitterness may also be related to off-flavour compounds from Maillard reactions, such as furan and caramel compounds.14
E Kayitesi et al.
Darker colour development was observed in FHMSP and DHMSP from heated beans and was confirmed by use of a colour meter. This may be due to non-enzymatic browning from Maillard-type reaction products developed during heating.15 A dark colour of marama flour from heated beans was reported by Maruatona.5 FUMSP and DUMSP from unheated beans were lighter in colour. The lighter colour of the marama/sorghum composite porridges from unheated beans was appreciated by most of the consumers in their comments as an improvement to the normal brownish colour of sorghum porridge. Light coloured sorghum porridges from white sorghum are usually preferred by consumers.16 The marama/sorghum composite porridges and TBTP were less viscous than the sorghum porridge. A reduction in viscosity of the composite porridges was identified by the descriptive sensory panel and confirmed by pasting measurements. A similar reduction in hot peak viscosity of sorghum porridge supplemented with peanut flour was reported by Singh and Singh.17 Texture measurements showed that the sorghum porridge was firmer than the composite porridges. Starch contributes greatly to the
5 4
DHMSP
A
PCA 2 : 24.34%
3 2 1
FHMSP
0
SOGP
DUMSPTBTP
-1 -2 -3
FUMSP
-4 -5 -5 -4 -3 -2 -1
0
1
2
3
4
5
6
7
8
PCA 1 52.15% 1.0 Bitter aftertaste Roasted nut aroma Sour aftertaste Roasted nut flavour
B
PCA 2 : 24.34%
0.5 aroma strengthBitter Overall Overall flavour strength
Salty 0.0 Sweet aromatic nutty aftertaste Oily mouth feel Buttery/creamy flavour
Viscosity Cooked sorghum aroma Cooked sorghum aftertaste Cooked sorghum flavour
-0.5 Boiled nut flavour Coarseness Boiled nut aroma -1.0 -1.0
-0.5
0.0 PCA 1 : 52.15%
0.5
1.0
2130
Figure 3. Principal component analysis of porridges: SOGP, sorghum porridge; FUMSP, full-fat unheated marama/sorghum porridge; FHMSP, full-fat heated marama/sorghum porridge; DUMSP, defatted unheated marama/sorghum porridge; DHMSP, defatted heated marama/sorghum porridge; TBTP, Tsabotlhe (sorghum/soya porridge) commercial product from Botswana. (A) Plot of first two principal component scores of porridges. (B) Plot of first two principal component loading projections of sensory attributes. (C) Plot of first and third principal component scores of porridges. (D) Plot of first and third principal component loading projections of sensory attributes.
www.interscience.wiley.com/jsfa
c 2010 Society of Chemical Industry
J Sci Food Agric 2010; 90: 2124–2132
Sensory quality of marama/sorghum composite porridges
www.soci.org
4 3
C
DUMSP PCA 3 : 12.80%
2 1
FUMSPDHMSP
0
SOGP FHMSP
-1
TBTP
-2 -3 -4
-5
-4
-3
-2
2 3 0 1 PCA 1: 52.15%
-1
4
5
7
6
8
1.0 D Bitter
PCA 3 : 12.80%
0.5 Bitter aftertaste Boiled nut flavour Sweet aromatic nutty aftertaste Sour aftertaste 0.0 Oily mouth feel Buttery/creamy flavour Roasted nut flavour Roasted nut aromaBoiled nut aroma
Viscosity Cooked sorghum flavour Cooked sorghum aroma Cooked sorghum aftertaste Overall aroma strength
Coaseness
-0.5
Salty Overall flavour strength -1.0
-1.0
-0.5
0.0 PCA 1 : 52.15%
0.5
1.0
Figure 3. (Continued).
J Sci Food Agric 2010; 90: 2124–2132
Most consumers that participated in the consumer acceptability test recommended the sorghum porridge for its firmness. The porridges made with full-fat marama flour were coarser than the other porridges. Marama bean seeds have a high fat content, and milling the beans resulted in a paste-like meal, making it difficult to obtain flour with homogeneous particles.
CONCLUSIONS The sensory attributes of sorghum porridges supplemented with marama flour were clearly different from those of traditional sorghum porridge. The sensory properties of the porridges affected their acceptability. The porridges prepared with marama flour from heated beans had a roasted nut aroma and flavour. The composite porridges were less viscous than the sorghum porridge (SOGP). SOGP and FHMSP were more acceptable. This is because SOGP was familiar to the consumers. FHMSP had rich, buttery and creamy flavours contributed by fat and sweet aromatic nut
c 2010 Society of Chemical Industry
www.interscience.wiley.com/jsfa
2131
textural properties of sorghum porridge. Marama flour has no starch,18 so introduction of marama flour decreases the starch levels in composite porridges. This may explain the reduction in viscosity and firmness of the porridges. Furthermore, high pasting temperatures and low paste viscosities have been reported in composite porridges with high lipid content.20 Starch molecules form helical inclusion lipid–amylose complexes.19,20 As a result, the composite porridge becomes less firm.21 Cereal-based porridges are the first solid foods to be introduced in a child’s diet in most developing countries. Their energy density is often low owing to the large volumes of water added during preparation to achieve a thin, drinkable consistency for infants.22 It has been reported that higher viscosity of porridges constrains the amount that can be consumed by infants.22 Combining sorghum meal with marama flour would be recommended to obtain a less viscous porridge with high energy density. This could be suitable for children in places where both marama and sorghum grow. Firmness of the sorghum porridge may have contributed to its acceptability.
www.soci.org 9 8
Hedonic rating
7
6.6c
6
5.8bc 5.0ab
5
5.4b 5.0ab
4.7a
4 3 2 1 SOGP
TBTP
FHMSP FUMSP
DHMSP DUMSP
Type of porridge
Figure 4. Consumer acceptability (1 = dislike extremely, 5 = neither like nor dislike, 9 = like extremely) of porridges: SOGP, sorghum porridge, TBTP, Tsabotlhe (sorghum/soya porridge) commercial product from Botswana; FHMSP, full-fat heated marama/sorghum porridge; FUMSP, full-fat unheated marama/sorghum porridge; DHMSP, defatted heated marama/sorghum porridge; DUMSP, defatted unheated marama/sorghum porridge. Mean values with different letters differ significantly (P < 0.05). The lighter shaded area is the higher percentile and the bottom represents the value above which 75% of the ratings fell. The darker shaded area is the lower percentile and the top represents the value above which 25% of the ratings fell. The median is the border between the two shaded areas.
flavours due to dry heating. Combining sorghum meal with full-fat marama bean flour has the potential to provide acceptable sensory appeal and could improve the nutritional status of persons using sorghum as a staple.
ACKNOWLEDGEMENT Financial support from the European Commission (Sixth Framework Program, MARAMAII Project) is gratefully acknowledged.
REFERENCES 1 Hartley ML, Tshamekang E and Thomas SM, Functional heterostyly in Tylosema esculentum (Caesalpiniodeae). Ann Bot 89:67–76 (2002). 2 Amarteifio JO and Moholo D, The chemical composition of four legumes consumed in Botswana. J Food Compos Anal 11:329–332 (1998). 3 Mmonatau Y, Flour from the marama bean: composition and sensory properties in a Botswana perspective. MScDissertation, Stellenbosch University, Cape Town (2005).
E Kayitesi et al.
4 Liener IE, Trypsin inhibitors: concern for human nutrition or not? J Nutr 116:920–923 (1986). 5 Maruatona G, Physico-chemical, nutritional and functional properties of defatted marama bean flour. MSc (Food Science) Dissertation, University of Pretoria (2008). 6 Kebakile MM, Rooney LW, de Kock HL and Taylor JRN, Effects of sorghum type and milling process on the sensory characteristics of sorghum porridge. J Cereal Chem 85:307–313 (2008). 7 Einstein MA, Descriptive techniques and their hybridization, in Sensory Science Theory and Applications in Foods, ed. by Lawless HT and Klein BP. Marcel Dekker, New York, NY, pp. 317–338 (1991). 8 Peryam DR and Pilgrim FT, Hedonic scale method of measuring food preferences. Food Technol 11:9–14 (1957). 9 Newport Scientific, RVA general article on paste viscosity. [Online]. (1998). Available: http://www.newport.com.au [20 April 2009]. 10 Ohmes RL, Marshall RT and Heymann H, Sensory and physical properties of ice creams containing milk fat or fat replacers. J Diary Sci 81:1222–1228 (1998). 11 Sacchetti G, Pinnavala GG, Guidoline E and Rosa DM, Effect of extrusion temperature and feed composition on the functional, physical and sensory properties of chestnut and rice flour based snack-like products. Food Res Int 37:527–534 (2004). 12 Guichard E, Interactions between flavour compounds and food ingredients and their influence of flavour perception. Food Rev Int 18:49–70 (2002). 13 Drewnowski A and Gomez-Carneros C, Bitter taste, phytonutrients, and the consumer. Am J Clin Nutr 72:1424–1435 (2000). 14 Bemiller JN and Whistler RL, Carbohydrates, in Food Chemistry, ed. by Fennema OR. Marcel Dekker, New York, NY, pp. 145–270 (1996). 15 Martins SIFS, Jongen WMF and Vanboekel MA, A review of Maillard reaction in food and implication of kinetic modelling. Trends Food Sci Technnol 11:364–373 (2001). 16 Aboubacar A, Kirleis AW and Oumarou M, Important sensory attributes affecting consumer acceptance of sorghum porridge in West Africa as related to quality tests. J Cereal Sci 30:217–225 (1999). 17 Singh U and Singh B, Functional properties of sorghum–peanut composite flour. Cereal Chem 68:460–463 (1991). 18 Van Zyl L, Microstructure of marama bean (Tylosema esculentum and Tylosema fassoglense) cotyledons and seed coats. EU Marama II Interim Report, University of Pretoria (2007). 19 Morrison WR and Boyd RN, Organic Chemistry (4th edn). Allyn and Bacon, Newton, MS, pp. 10–30 (1983). 20 Mishra S and Rai T, Morphology and functional properties of corn, potato and tapioca starches. Food Hydrocolloids 20:557–566 (2006). ˆ in Agriculture 21 Fliedel G, Appraisal of sorghum quality for making to, et Development, ed. by Saint Macary H and Jourda-Ruf CC. CIRAD, Montpellier, pp. 34–42 (1995). 22 Stephenson DM, Gardner JMM, Walker SP and Ashworth A, Weaningfood viscosity and energy density: their effects on ad libitum consumption and energy intake in Jamaican children. J Am Clin Nutr 60:465–469 (1994).
2132 www.interscience.wiley.com/jsfa
c 2010 Society of Chemical Industry
J Sci Food Agric 2010; 90: 2124–2132