ARTICLE IN PRESS

Rheological, texture and sensory properties of low-fat mayonnaise with different fat mimetics
H. Liua,b, X.M. Xua,b,, Sh.D. Guob
b

Key Laboratory of Food Science and Safety, Ministry of Education, China School of Food Science and Technology, Southern Yangtze University, Jiangsu, Wuxi 214036, China

Abstract Application of whey protein isolate and low-methoxy pectin-based fat mimetics in mayonnaise was studied. Fat was partially substituted by different fat mimetics at levels of 50%, respectively, which the fat mimetics were referred to as PFM1 (microparticulate pectin gel), PFM2 (pectin weak-gel), and PFM3 (microparticulated combination of WPI and pectin). The full fat (Ff) (100% oil) mayonnaise without fat mimetic was used as a control experiment. Physicochemical, rheological, texture analysis, and sensory evaluation of the Ff and low fat (Lf) mayonnaises were performed. The results indicated that all Lf mayonnaises had signicantly lower energy content, but higher water content than their Ff counterpart. In terms of texture, the formulation with pectin weak-gel as fat mimetic showed similar texture values as those of the Ff sample. Both Ff and Lf mayonnaises exhibited thixothopic shear thinning behaviour under steady shear tests and were rheologically classied as weak gels under small amplitude oscillatory shear tests. Sensory evaluation demonstrated that mayonnaises substituted with low-methoxy pectin were acceptable. This study shows good potential for pectin weakgel and microparticulated pectin gel to be used as a fat mimetic in mayonnaise.

Keywords: Mayonnaise; Low fat; Fat mimetics; Microparticulate; Whey protein isolate; Pectin gel; Weak-gel

1. Introduction Mayonnaise is a kind of semi-solid oil-in-water emulsion. It is traditionally prepared by carefully mixing a mixture of egg yolk, vinegar, oil, and spices (especially mustard) to maintain closely packed foam of oil droplets; it may also include salt, sugar or sweeteners, and other optional ingredients. The emulsion is formed by slowly blending oil with a pre-mix that consists of egg yolk, vinegar, and mustard because mixing the oil and aqueous phase at once will result in formation of a water-in-oil emulsion. It is probably one of the oldest and most widely used sauces in the world today. Traditional mayonnaise is
Corresponding author. School of Food Science and Technology, Southern Yangtze University, 214036 Wuxi, Jiangsu Province, China. Tel.: +86 510 85863533; fax: +86 510 85879711. E-mail address: xmxu@sytu.edu.cn (X.M. Xu).

an oil-in-water emulsion containing 7080% fat (Depree & Savage, 2001). Generally, mayonnaise stability is dependent on several factors such as amount of oil, amount of egg yolk, viscosity, relative volume of oil phase to aqueous phase, method of mixing, water quality, and temperature (Harrison & Cunningham, 1985). Over the past decade, the consumption of low-fat (Lf) food products has become more than just a trend. In view of the general consensus that the amount and type of fat consumed is of importance to the aetiology of several chronic diseases (e.g. obesity, cardiovascular diseases, cancer), it is not surprising that consumers more readily adhere to nutritional guidelines concerning fat consumption. Largely inuenced by healthrelated concerns, there has been pressure on the food industry to reduce the amount of fat, sugar, cholesterol, salt and certain additives in the diet. Food manufacturers have responded to consumer demand and there has been rapid market growth of products with a healthy image. In

ARTICLE IN PRESS
addition to nutrition, fat inuences rheological properties and sensory characteristics of foods such as avour, mouthfeel and texture. These sensory properties are very hard to reproduce in formulations without fats. Therefore it is rather difcult to imitate traditional product quality when preparing Lf foods. There is the possibility of choosing particular fat substitutes in specic quantities in order to have a product with a texture close to that of traditional mayonnaise. Thus to establish the formulation of the Lf products, food technologists have focused their efforts essentially on fat replacers. From a physical point of view, mayonnaise is an oil-in-water emulsion (O/W) and therefore to produce a dietetic mayonnaise it is necessary to decrease the dispersed phase and to increase the water content. Some fat mimetics such as modied starch (Murphy, 1999), inulin (James, 1998), pectin (Pedersen & Christian, 1997) and microcrystalline cellulose (Chouard, 2005; Grodzka, Maciejec, & Krygier, 2005), carrageenan (Trueck, 1997), some thickeners (Wendin et al., 1997) were generally used to stabilize the emulsion and to increase the viscosity of light mayonnaise. The rheology of mayonnaise was investigated by several authors because of its importance for the choice of formulation, process conditions and quality control. Mayonnaise shows a yield stress, a pseudoplastic behaviour and time dependent characteristics (Peressini, Sensidoni, & Cindio, 1998; Batista, Raymundo, Sousa, & Empis, 2006). Steady shear, oscillatory shear measurements were carried out on mayonnaise samples by many authors. For the steady shear measurement, the power law, carreau, HerschelBulkley and Casson models have been widely used to describe the ow properties of mayonnaise and salad dressing (Peressini et al., 1998; Guilmineau & Kulozik, 2007; Batista et al., 2006). Mayonnaises are oilin-water emulsions with a texture that is particularly appreciated by consumers. The actual nutritional trend towards low-calorie foods has increased the interest in fat substitutes without altering the consistency of the product. From this point of view rheological properties may give a quantitative contribution to texture characterization and control when using different formulations. Texture analysis of mayonnaise was done recently (Batista et al., 2006; Worrasinchai , Suphantharika, Pinjai, & Jamnong, 2006). The result of texture analysis had better discriminating ability to compare emulsions. As introduced in some patents, a series fat mimetics based on starch, protein and gum were used in Lf mayonnaise or salad dressing (Bakal, Galbreath, & Cash, 1992; Pedersen & Christian, 1997; James, 1998; Aluko & McIntosh, 2005). But the application effects were not study systemic. In this paper, the author used new technics to prepare the fat mimetics based on WPI and pectin and investigate the effects of three fat mimetics prepared from WPI/pectin and pectin, respectively, substitution for fat in mayonnaise formulations on rheological, texture and sensory properties of the Lf mayonnaises. The full fat (Ff) formulation was used as reference. 2. Materials and methods 2.1. Materials Whey protein isolate, which contained more than 90% protein, was a gift from Pacic Food Ingredient Tech. Inc., China. LM pectin was sample from Danisco A/S., Denmark. Other ingredients as composition of the model mayonnaise such as soybean oil, egg, acetic acid solution, sugar and salt were all purchased from a local supermarket. The recipes and preparation method of the mayonnaise were modied from the introduction of Xiong, Xie, and Edmondson (2000) and Worrasinchai et al. (2006), the recipes of the Ff and Lf mayonnaise are shown in Table 1. 2.2. Preparation of the fat mimetics There were three fat mimetics applied in this study. They were used in mayonnaise formulations referred as PFM1, PFM2, and PFM3. The PFM1 was microparticulated pectin gel. Firstly, the pectin was dissolved in water and then forming pectin gel with addition of the calcium ion. Chopping the gel to form a processable mass and shearing the coarse particles to form non-spheroidally shaped gel particles having the substantially smooth organoleptic character of an oil emulsion. The PFM2 was formed by mixing the pectin with water and interacted with calcium ion to form a weak-gel. PFM3 was the combination of WPI and low-methoxy pectin. The interaction of the WPI and pectin strengthened the gelation properties and improve the water holding capacity of WPI and pectin, respectively, but the data not shown here. The combination of the WPI and pectin was then microparticulated for further used in Lf mayonnaise. Due to the privacy, detailed introductions of the preparation of the fat mimetics were omitted. The percentage moisture of the PFM1, PFM2, and PFM3 was approximately 97, 98, and 90, respectively.

Table 1 Percentage recipes of the mayonnaise (wt, %) Ffa Soybean oil Egg yolk Fat mimetic Vinegar solution Mustard Salt Sugar
a b

LfP1b 40 8 40 8 1 2 1

LfP2c 40 8 40 8 1 2 1

LfWd 40 8 40 8 1 2 1

80 8 8 1 2 1

Full fat mayonnaise. Low fat mayonnaise with microparticulate pectin gel (PFM1) as fat mimetic. c Low fat mayonnaise with pectin weak-gel (PFM2) as fat mimetic. d Low fat mayonnaise with microparticulate combination of WPI and pectin (PFM3) as fat mimetic.

ARTICLE IN PRESS
2.3. Preparation of the mayonnaise The water phase was rst prepared by mixing all of the ingredients apart from the oil in a 1 l plastic beaker. The oil was then carefully mixed with the water phase. In this study, for the reduced fat (Lf) mayonnaises, soybean oil was replaced by fat mimetics at levels of 50% of total oil used. The fat mimetics were microparticulated gel from whey protein isolate and pectin interaction reactant, microparticulated pectin gel solely and pectin solgel solution referred as LfW, LfP1 and LfP2 formulations, respectively. Mayonnaise were aseptically transferred to sterile 500 ml glass beakers sealed with polypropylene lm and left several days at 4 1C before testing. 2.4. Composition analysis The amounts of moisture and ash in the mayonnaise samples were measured by AOAC ofcial method 926.08 (1995), 935.42 (1995), respectively. Total protein and total fat content of the mayonnaise were determined using the Kjeldahl, and the modied Mojonnier method, respectively (Marshall, 1992). The protein content of cheeses was calculated by multiplying the total nitrogen content by 6.38. Carbohydrates were determined by subtracting the sum of moisture, protein, fat, and ash percentages from 100%. 2.5. Rheology analysis The rheological measurements were performed in a rheometer (TA Instruments AR-1000, UK). The mayonnaise ow properties at 25 1C was analysed using a parallel stainless steel plate having a diameter of 40 mm. A thixotropic loop measurement was carried out by rst increasing the shear rate logarithmically from 0 to 150 s1, then maintaining it at 150 s1 for 240 s, and nally decreasing it logarithmically back to 0 s1. In order to perform a quantitative comparison of materials, it is generally required tting the experimental data to some forms of best-t mathematical equation or model. The model is used to predict the viscosity of a ow as a function of shear stress or shear rate, it is named also constitutive equation of a ow. There were some useful ow models introduced to use for tting the ow curve of mayonnaise or salad dressing as follows. (A) Carreau model (Batista et al., 2006): Z0 Z , (1) 1 _ =_ c 2 s g g where Z0 is the limiting viscosity for the rst Newtonian _ region, gc is the critical shear rate for the onset of the shear thinning behaviour and s is a parameter related to the slope of this region. (B) HerschelBulkley equation (Guilmineau & Kulozik, 2007): This model incorporates the elements of the Mayonnaise texture measurements were carried out with the TA.XT2i Texture Analyser (Stable Micro Systems Ltd, Surrey, UK) with a 25 kg load cell. Back extrusion cell with 35 mm diameter compression disc was used. The samples were carefully scooped into acrylic cylindrical containers (50 mm internal diameter and 75 mm height) to a depth of 50 mm. One cycle was applied, at a constant crosshead velocity of 1 mm/s, to a sample depth of 40 mm, and then returned. From the resulting forcetime curve, the values for texture attributes, i.e. rmness, consistency, cohesiveness, viscosity were obtained using the Texture Expert for Window Version 1.22 equipment software. Firmness was the maximum force as the test cell penetrated into the sample as introduced in Worrasinchai et al.s (2006) article. When a 30 g surface trigger is attained (i.e. the point at which the disc lower surface is in full contact with the product) the disc proceeds to penetrate to a depth of 40 mm (or other specied distance). At this point (most likely to be the maximum force), the probe returns to its original position. The peak or maximum force is taken as a measurement of rmnessthe higher the value the more rm is the sample. The area of the curve up to this point is taken as a measurement of consistencythe higher the value the thicker the consistency of the sample. The negative region of the graph, produced on probe return, is as a result of the weight of sample which is lifted primarily on the upper surface of the disc on return, i.e. due Newtonian, Power law and Bingham equations _ t ty K gn , (2)

where Newtonian yield stress ty 0, n 1, Power law yield stress ty 0, n power law index, and Bingham yield stress ty value, n 1. In this equation, t is the shear stress (Pa), ty the yield _ stress (Pa), g the shear rate (s1), K the consistency index (Pa sn) and n the ow index. It can often be used as a general purpose model for shear thinning materials. An appropriate model is selected to obtain the inrheology. However, when the author used Carreau model to tting the experiments data, it was found that the standard error was higher than 100 and the standard error should better less than 20. So in this paper, the HerschelBulkley equation model was used to compare the different samples. Oscillatory tests (mechanical spectra) were performed in a frequency range of 0.160 Hz, using strain values comprised in the linear viscoelastic region for each sample (0.5%). Data were collected and rheological parameters were calculated using TA instrument software programme. The recorded data included the components of shear modulus G0 (storage modulus) and the viscous component G00 (loss modulus). The components of tan(delta) and complex viscosity (jZ j) were also measured. Then the curves of G0 , G00 , jZ j and tan(delta) vs. frequency were obtained. 2.6. Texture analysis

ARTICLE IN PRESS
to back extrusion and hence gives again an indication of consistency/resistance to ow off the disc. The maximum negative force is taken as an indication of the stickiness (or may in this case be referred to as cohesiveness) of the samplethe more negative the value the more sticky or cohesive is the sample. The area of the negative region of the curve is often referred to as the work of adhesion the higher the value the more resistant to withdrawal the sample is which is perhaps an indication again of the cohesiveness and also consistency/viscosity of the sample. 2.7. Particle size measurement 2.10. Statistical analysis Mayonnaise samples (0.04 g) were diluted with 150 ml 0.1% sodium dodecyl sulfate (SDS) solution, stirred gently by a spatula until completely dispersed. Samples were analysed by a particle size analyser (Mastersizer, 2000, Malvern Instrument Ltd, Worcestershire, UK). The sample solution was dispersed in distilled water at 1500 rpm until an obscuration rate of 0.20.5% was obtained. Background and sample were measured for 10 s. Optical properties of the sample were dened as follows: refractive index 1.460 and absorption 0.00. Droplet size measurements are reported as the volume-weighted mean diameter: P 4 P 3 d 4;3 ni d i = ni d i where ni is the number of droplets of diameter di. Each sample was measured in triplicate (Worrasinchai et al., 2006). 2.8. Optical microscope observation The glassy at was coated with mayonnaise sample and placed on the stage of Inverted Microscope Eclipse TS100 (Nikon, Japan). Then adjusted the focus knob to get clearly view eld. The pictures of the mayonnaise microstructure were obtained by a digital camera connected with the microscope. 2.9. Sensory analysis Sensory evaluation was conducted on the mayonnaise samples after one-day storage at room temperature. Initially, panelists were trained in 2 h sessions prior to The composition analysis and caloric values of the Lf mayonnaises containing fat mimetics, and the Ff mayonnaise (control) are listed in Table 2. The moisture content increased with addition of fat mimetics due to very high moisture content of fat mimetics preparation, which were typical characteristic of carbohydrate or protein-based fat replacers. Because of the difference of the oil content addition to Ff and Lf samples, fat content in Lf samples were much lower than the Ff samples. Carbohydrate contents in Lf samples were higher than the Ff samples for pectin base fat mimetics were added in the Lf mayonnaise. The caloric values of the Lf formulations were signicantly (Pp0.05) reduced. Besides water as a main component of A one-way analysis of variance for the data for composition analysis, textural analysis and sensory values were carried out to determine the signicance of the individual differences. Signicant means were compared using F test and standard deviations for mean values of chemical analysis were also calculated. Simple correlations were performed between the rheology, textural, sensory analysis of the mayonnaise samples. All the statistical analysis was conducted using the Matlab (Version 6.5) commercial statistical package. 3. Results and discussion 3.1. Chemical composition and caloric values evaluation to be familiar with attributes and scaling procedures of mayonnaise samples under study. Sensory characteristics: appearance, colour, odour, texture, taste, and overall acceptability were evaluated by 30 trained panel on 9-point hedonic scale, 1 the least, the lowest; 9 the most, the highest. All mayonnaise samples were coded with three-digit random numbers and presented to panelists on a tray in individual booths. Orders of serving were completely randomized. Water was provided between samples to cleanse the palate (Yazici & Akgun, 2004; Worrasinchai et al., 2006)

Table 2 Chemical composition analysis (%, w/w) and caloric values of mayonnaise samples Samples FfMb LfP1c LfP2d LfWe Moisture content 13.7470.12b 54.3670.26a 53.2170.42a 47.4970.34a Fat 82.5471.23a 41.3570.86b 42.4370.98b 43.8670.57b Carbohydrate 1.54b 2.03a 2.10a 2.00a Protein 1.3470.06b 1.3670.07b 1.3770.05b 5.4270.45a Ash 0.8470.02b 0.9070.05b 0.8970.03b 1.2370.11a Caloric valuesa 754.38711.43a 385.7178.50b 395.7579.38b 424.4277.33b

Assays were performed in triplicate. Mean GSD values followed by the same letter in each column are not signicant different at Pp0.05 by ANOVA and Tukeys test. a Caloric values (4 protein)+(9 fat)+(4 carbohydrate). b Full fat mayonnaise. c Low fat mayonnaise with microparticulate pectin gel (PFM1) as fat mimetic. d Low fat mayonnaise with pectin weak-gel (PFM2) as fat mimetic. e Low fat mayonnaise with microparticulate combination of WPI and pectin (PFM3) as fat mimetic.

ARTICLE IN PRESS
fat mimetics sample, the pectin itself is non-caloric because it is not digested or absorbed in the human digestive tract. Although WPI give 9 cal/g energy when digested by human, the content of the WPI was not so much, the caloric value of the LfW is lower than the Ff samples but little higher than the LfP1 and LfP2 samples. 3.2. Rheological characteristic Flow curves of the Ff and Lf mayonnaises are presented in Fig. 1. All mayonnaise samples exhibited thixotropic behaviour over the whole range of shear rate studied (0150 s1) in that segments of the down curves reached values of shear stress lower than those of the up curves at the same values of shear rate. The thixotropic characteristics of mayonnaise in the shear rate range of 0200 s1 were reported earlier (Stern, Valentova, & Pokorny, 2001). ` All investigated mayonnaise samples showed thixotropic shear thinning behaviour in that their ow properties depended on both shear rate and time. In concentrated emulsions, the droplets are close enough together to interact with each other which may lead to the formation of a three-dimensional network of aggregated droplets. As the shear rate is increased, the hydrodynamic forces cause aggregates to become deformed and eventually disrupted which results in a reduction in the viscosity (McClements, 1999). Thixotropic loop was obtained for each sample by successively increasing, maintaining and decreasing the shear rate. The surfaces of the area comprised within the thixotropic loop, referred to as thixotropy, are shown in Table 3. Thixotropy was larger in mayonnaise containing higher fat compared to those Lf samples. This increased level of thixotropy corresponds to a progressive breakdown of the products structure as the time of shear is increased (Abu-Jdayil, 2003). The thixotropy of Ff mayonnaise was about 9 times, 1.6 times and 4 times larger than LfP1, LfP2
Table 4 Model-tting ow equation parameters of mayonnaise samples Samples Ff LfP1 LfP2 LfW ty (Pa) 12.71 0.02479 8.706 9.591 K (Pa sn) 7.238 4.042 7.240 0.5733 n 0.4821 0.4698 0.4820 0.8161 Standard error 11.56 8.715 12.06 49.31 Table 3 Thixotropy of mayonnaise samples Samples Thixotropy (Pa/s) FfM 4399 LfP1 477.8 LfP2 2768 LfW 1115

100

80 shear stress(Pa)

60

40

20

20

40

60

80

100

120

140

160

shear rate(s-1)
Fig. 1. Flow curves of samples: thixotropy loop. --, Ff; --, LfP1; -m-, LfP2; -.-, LfW.

and LfW, respectively. This result implied that the structure of LfP2 was more similar with the Ff mayonnaise than other Lf samples, which could nd proof in Fig. 3. As Fig. 3 showed that, the Ff mayonnaise had uniform fat granules distributed in water system. The fat granules were less when fat was replaced by fat mimetics. In Lf mayonnaise with pectin weak-gel addition, the granules were pockety in egg-box structure. In Lf samples with microparticulated pectin and combination of WPI/pectin as fat mimetic, the particles were obviously found. Table 4 showed the model-tting ow equation parameters of different mayonnaise samples. The yield stress of the Ff mayonnaise was the largest one in all the mayonnaise samples. When used as the salad dressing, the yield stress was an important character for it must have properly ability to keep non-uid on the salad surface. The samples with pectin weak-gel and microparticulated combination WPI/pectin as fat mimetic had similar yield stress and lower than the Ff samples. The sample with microparticulated pectin gel had the lowest yield stress and the value was surprisingly too low. This might be due to the inapposite model-tting. Because when the container, which was lled with the four mayonnaises, was inverted, the samples did not ow which showed that they all had considerable yield stress. The consistency coefcient (K) values were markedly decreased for LfP1 and LfW samples and the K values of LfP2 and the Ff were similar. In terms of ow behaviour index (n), the Ff, LfP1, and LfP2 showed more non-Newtonian ow behaviour than LfW whose ow behaviour index was 0.8161 near to 1.0. Mechanical spectra of the Ff and Lf mayonnaises obtained by the frequency sweep measurement classied into that of weak gels because G0 was larger than G00 throughout the tested frequency range and both G0 and G00 hardly correlated with the frequency change (Fig. 2) and the separation of the two moduli (tan(delta)) is larger than 0.1 (Fig. 2). The previous studies also found that mayonnaise exhibited weak-gel characteristics in the frequency range of 0.110 Hz (Mancini, Montanari,

ARTICLE IN PRESS

A
10000

B
10000

1000 1000 G''(Pa) G'(Pa) 100

100

10 10 1 0.1 1 frequency (Hz) 10 0.1 1 frequency (Hz) 10

C
10000 1000 |n|(Pa.s) 100 10 1

D
1

tan(delta) 0.1 0.1 1 frequency (Hz) 10

0.1 0.1 1 frequency (Hz) 10

Fig. 2. Dynamic mechanical spectra of mayonnaise samples: (A) G0 , (B) G00 , (C) jZ j and (D) tan(delta) vs. frequency, respectively. --, Ff; --, LfP1; -m-, LfP2; -.-, LfW.

Peressini, & Fantozzi, 2002). In general, it is expected that emulsions with a greater fat content show higher values of G0 (Ma & Barbosa-Canovas, 1995). In this study, however, this effect was counterbalanced by the substitution of fat mimetics for fat in the formulation that strengthened the gel structure of mayonnaises. Loss tangent (tan(delta)) indicates whether elastic or viscous properties predominate in a sample. A small amplitude oscillatory experiment, carried within the linear viscoelastic region, on the other hand, has the advantage of minimizing destruction in the sample since little or no permanent structure breakdown occurs during the dynamic measurements. This approach allows a relationship between the results obtained and the actual structure of material to be drawn. In the oscillatory test all samples exhibited viscoelastic behaviours with a storage modulus (G0 ) greater than the loss modulus (G00 ). Generally speaking it is expected (Ma & Barbosa-Canovas, 1995) that emulsions with a greater fat content show higher values of G0 . Ff sample exhibited

the highest storage modulus. Based on the observation of Fig. 2 and Fig. 3, the samples that had more compact structure had the higher storage modulus. The LfP1 and LfP2 in spite of the same lower oil content, had higher G0 probably because the G0 reduction was counterbalanced by the weak-gel increase in formulation, that structure the material. It is useful also to report the tan(delta) for the tested samples to have a direct view on whether the different samples behave as liquid or solid. In the range of frequency tested, the Ff, LfP1, LfP2 showed solid-like for the tan(delta) values were less than 1.0, but the LfW had a liquid-like sample trend region because the tan(delta) value was close to 1.0 at high frequency. The LfP2 showed the lowest tan (delta) value and manifest it was more solid-like than others mayonnaise samples. From these results, it can be asserted that the compact packing of oil droplets into network was responsible for elastic properties and deformation resistance of the emulsion. The pectin weak-gel played good roles in strengthening the network structure.

ARTICLE IN PRESS

Fig. 3. Optical microscope of different mayonnaise formulations ( 100) scale bar 100 mm.

Table 5 Texture analysis of mayonnaise samples Samples Ff LfP1 LfP2 LfW Firmness (g) 346.4373.67a 256.5672.19b 349.3572.67a 257.6871.98b Consistency (g s) 3654.7879.34a 2785.3578.89b 3649.6279.63a 2878.5279.33b Cohesiveness (g) 198.4572.73a 120.4571.23b 193.6073.69a 139.3572.64b Viscosity (g s) 1672.5778.42a 983.5673.68b 1685.4179.24a 1002.3675.68b

3.3. Texture analysis All the texture parameters determined are shown in Table 5. The Ff and LfP2 were similar for rmness, consistency, cohesiveness and viscosity values and signicantly higher than those of the LfP1 and LfW. In mayonnaise, the large contact surface area between oil droplets leads to important friction force, which oppose to the free owing of the emulsion in a shear eld, hence increasing its viscosity. A decrease of oil droplet diameter leads to a greater contact surface area between droplets, and therefore to an increased viscosity. The viscosity of the samples can partially but not all reect the texture analysis parameters. So the hypothesis could be given that the oil phase droplet diameter of Ff and LfP2 were smaller than the other two samples but the results of particle size analysis (Fig. 4) showed that the Lf mayonnaise with pectin weak-gel as fat mimetic had the largest particle size. This

90 80 70 particle size (m) 60 50 40 30 20 10 0 Ff LfP1 sample

Fig. 4. The particle size of the different samples. , particle size.

LfP2

LfW

ARTICLE IN PRESS

Fig. 5. Hypothetic schematic representation of fat mimetics interacting with oil/water emulsion.

may be due to the fact that the samples were hardly dispersed in solution and the induced results were higher than its intrinsic particle size. Based upon the results, the author made hypothetic schematic representation of fat mimetics interacting with oil/water emulsion as shown in Fig. 5. The contribution of pectin weak-gel to the viscosity of the mayonnaise may be from the tangly interaction between the pectin gel three-dimension structures with oil granules. And the particle size of fat mimetics was important. The better roles as fat mimetics need the microparticulated pectin gel or microparticulated combination of WPI/pectin had the similar particle size of oil granule. 3.4. Sensory evaluation Sensory evaluation scores of the Ff and Lf mayonnaises are shown in Fig. 6. The appearance score of LfW signicantly (Pp0.05) lower than the others may due to the WPI/pectin base fat mimetic effected the appearance of the mayonnaise. The colour of the Ff sample was better than the Lf samples. The higher moisture content in the Lf mayonnaise diluted the colour of the samples, and in the mean time the Lf sample with microparticulated combination of WPI/pectin obtained the lowest colour score which might be due to the white colour of the fat mimetic resulting in the colour of the LfW off-white. The pectin base fat mimetic, on the other hand, did not signicantly inuence the odour attribute. The odour of the WPI itself affected the samples odour and decreased the odour score. The Ff and Lf with pectin weak-gel samples showed no statistical difference in texture and taste scores, whereas the other two Lf samples gave lower scores for these attributes. The texture attribute was well correlated with the data derived from instrumental analysis using the texture analyser (Table 5). But the correlation of the sensory analysis and rheological analysis was chaos. One reason for this could be the measurement geometry used is not representative of the action of the mouth. In further study, tribology and thin lm rheological techniques (Zhang, Matar, & Craster, 2003; Narsimhan & Wang, 2005; Vicente, Stokes, & Spikes, 2006) might be used to compare the instrumental analysis and sensory analysis. The overall

12 10 8 6 4 2 0 Appearance Colour Odour Texture Taste Acceptability sensory parameters

Fig. 6. Sensory evaluation of mayonnaise samples. , LfP2; , LfW. , Ff; , LfP1;

acceptability scores were signicantly lower when the substitution was microparticulated pectin gel or microparticulated WPI/pectin. This is mainly contributed by the performance of appearance and colour attributes. In this study, the overall acceptability score of one commercial mayonnaise evaluated by the same sensory panel was approximately 6 (data not shown). Therefore, it is reasonable that the sensory attributes with scores higher than 6 are considered acceptable when compared with the controls. Thus, the Lf mayonnaises with microparticulated pectin gel and pectin weak-gel were judged to be sensorial acceptable. 4. Conclusions From the results of the present work, it can be concluded that pectin solgel and microparticulate pectin gel played a better role as a fat mimetic which resulted in the Lf mayonnaises with lower calorie than their Ff counterpart but similar texture characteristic properties and had a better acceptability than the WPI/pectin base fat mimetic added to Lf mayonnaise.

sensory score

ARTICLE IN PRESS
References
Abu-Jdayil, B. (2003). Modelling the time dependant rheological behaviour of semi-solid foodstuffs. Journal of Food Engineering, 57, 97102. Aluko, R. E., & McIntosh, T. (2005). Limited enzymatic proteolysis increases the level of incorporation of canola proteins into mayonnaise. Innovative Food Science and Emerging Technologies, 6, 195202. AOAC Ofcial Method 926.08 (1995). Moisture in cheese (16th ed.) (Vol. II, p. 58). Ofcial methods of analysis of AOAC International. Gaithersburg, MD: AOAC International. AOAC Ofcial Method 935.42 (1995). Ash of cheese (16th ed.) (Vol. II, p. 59). Ofcial methods of analysis of AOAC International. Gaithersburg, MD: AOAC International. Bakal, A.I., Galbreath, T., & Cash, P.A. (1992). Fat free and low fat mayonnaise-like dressings. United States of American Patent. No. 651652. Batista, A. P., Raymundo, A., Sousa, I., & Empis, J. (2006). Rheological characterization of coloured oil-in-water food emulsions with lutein and phycocyanin added to the oil and aqueous phases. Food Hydrocolloids, 20, 4452. Chouard, G. (2005). 100% indulgence 0% guiltbe in phase, bring connectivity to your low-fat mayonnaise. Innovations in Food Technology, 29, 98100. Depree, J. A., & Savage, G. P. (2001). Physical and avour stability of mayonnaise. Trends in Food Science and Technology, 12, 157163. Grodzka, K., Maciejec, A., & Krygier, K. (2005). Attempts to apply microcrystalline cellulose as a fat replacer in low fat mayonnaise emulsions. Zywnosc, 12, 5261. Guilmineau, F., & Kulozik, U. (2007). Inuence of a thermal treatment on the functionality of hens egg yolk in mayonnaise. Journal of Food Engineering, 78, 648654. Harrison, L. J., & Cunningham, F. E. (1985). Factors inuencing the quality of mayonnaise. Journal of Food Quality, 8, 120. James, S. M. (1998). Method for producing fat-free and low-fat viscous dressings using inulin. United States of American Patent No. 5721004. Ma, L., & Barbosa-Canovas, G. V. (1995). Rheological characterization of mayonnaise. Part II: Flow and viscoelastic properties at different oil and xanthan gum concentrations. Journal of Food Engineering, 25, 409425. Mancini, F., Montanari, L., Peressini, D., & Fantozzi, P. (2002). Inuence of alginate concentration and molecular weight on functional properties of mayonnaise. Lebensmittel-Wissenschaft Und-Technologie, 35, 517525. Marshall, R. T. (1992). Standard methods for the examination of dairy products. Washington, DC: American Public Health Association. McClements, D. J. (1999). Food emulsions; principles, practice, and techniques. Boca Raton, FL: CRC Press. Murphy, P. (1999). Low fat developments with speciality starches. Food Technology International, 22, 24. Narsimhan, G., & Wang, Z. (2005). Stability of thin stagnant lm on a solid surface with a viscoelastic airliquid interface. Journal of Colloid and Interface Science, 291, 296302. Pedersen, A., & Christian, H. (1997). No and low fat mayonnaise compositions. United States of American Patent No. 5641533. Peressini, D., Sensidoni, A., & Cindio, B. d. (1998). Rheological characterization of traditional and light mayonnaises. Journal of Food Engineering, 35, 409417. Stern, P., Valentova, H., & Pokorny, J. (2001). Rheological properties and ` sensory texture of mayonnaise. European Journal of Lipid Science and Technology, 103, 2328. Trueck, H. U. (1997). A mayonnaise-like product. European Patent Application No. EP 0768042A1. Vicente, J. de., Stokes, J. R., & Spikes, H. A. (2006). Soft lubrication of model hydrocolloids. Food Hydrocolloids, 20, 483491. Wendin, K., et al. (1997). Low-fat mayonnaise: Inuences of fat content, aroma compounds and thickeners. Food Hydrocolloids, 11, 8799. Worrasinchai, S., Suphantharika, M., Pinjai, S., & Jamnong, P. (2006). b-Glucan prepared from spent brewers yeast as a fat replacer in mayonnaise. Food Hydrocolloids, 20, 6878. Xiong, R., Xie, G., & Edmondson, A. S. (2000). Modelling the pH of mayonnaise by the ratio of egg to vinegar. Food Control, 11, 4956. Yazici, F., & Akgun, A. (2004). Effect of some protein based fat replacers on physical, chemical, textural, and sensory properties of strained yoghurt. Journal of Food Engineering, 62, 245254. Zhang, Y. L., Matar, O. K., & Craster, R. V. (2003). Analysis of tear lm rupture: Effect of non-Newtonian rheology. Journal of Colloid and Interface Science, 262, 130148.