Particle Size Distribution of Natural Peanut Butter and its Dynamic Rheological Properties

Transcription

1 Proceedings of the 204 International Conference on Food Properties (ICFP204) Kuala Lumpur, Malaysia, January 24 26, 204 Particle Size Distribution of Natural Peanut Butter and its Dynamic Rheological Properties Mohd Rozalli, N. H. a, Chin, N. L. b*, and Yusof, Y. A b. a Department of Food Technology, School of Industrial Technology, Universiti Sains Malaysia, 800 Penang, Malaysia b Department of Process and Food Engineering, Faculty of Engineering, Universiti Putra Malaysia, UPM Serdang, Selangor, Malaysia * Corresponding author. Tel.: ; fax: address: chinnl@upm.edu.my (Chin, N. L) Abstract This study compares the dynamic rheological properties of natural peanut butter with commercial peanut butter at 25 C. The natural peanut butter was produced using ultra-high speed grinding (~20000 rpm) at different grinding times (2-5 minutes) from peanuts of China and India. Multimodal particle size distribution (PSD) was observed for all the samples. The linear viscoelastic region (LVR) obtained from stress sweep test of the peanut butter produced at 2-3 minutes fell within the LVR region of commercial peanut butter of 0.- Pa. Longer grinding time (3.5-5 minutes) produced a shorter and lower LVR of Pa. The storage modulus, G is an increasing function of PSD. All peanut butter samples exhibited elastic properties. Keywords: peanut butter, particle size distribution, rheology, structure, viscoelastic.0 Introduction Peanut butter is a spreadable product made from roasted peanuts that are ground into paste. The term, peanut butter is used for a product consisting of minimum 90% of peanuts while less than that is considered as peanut spread. In current peanut butter manufacturing, peanuts have to undergo two steps of grinding as described by Hlavacek and Freeman (968) in order to obtain the finest peanut butter []. The initial milling reduces the nuts into medium grind and the second milling further produces an ultrafine particle size of peanut butter. The peanuts have to be grinded in two different grinders due to the different plates clearance required at each stage. Ultimately, an ultrafine particle size is produced i.e. 98% less than 73.7 µm diameter (d 4,3 ) with a maximum size less than 254 µm using a very high speed comminutor operating at 9600 rpm []. In other study, the particle size of smooth peanut butter product examined under light microscopy showed that the volume average diameters (d 4,3 ) is 6.6 µm [2]. Whilst differences in particle size affects peanut butter texture, different grinding method also yield product of different texture as reported by Sihsobhon et al (2008) [3]. Food products with same particle size but differ in particle size distribution (PSD) are known to yield product with dissimilar texture and structure. Broader PSD of the whole jaboticaba pulp shows different rheological properties than the reconstituted pulp even though at the same particle size [4]. The importance of structure rheology interrelationship had been highlighted by many authors [5-8]. The particle size analysis and rheological characterization of food products are important in their formulation, processing, transportation and storage, particularly for food emulsions and suspensions [8]. Peanut butter is a kind of colloidal suspension consisting solid peanut particles in peanut oil [2]. Both rheological and structural data should lead to understanding the interrelations between them which further enhances the food quality through invention of foods with desirable structures and rheological behaviour [6]. Controlling particle size distribution (PSD) was proved to be the most prominent factor to obtain a desirable product ([9]. The small amplitude oscillatory shear (SAOS) test within the LVR had been extensively studied for better understanding of the relationships between structural and rheological properties of food material such as mozzarella cheese [0], sesame paste [8], apple puree [] and tamarind seed gum in aqueous solution [2]. Tunick (200)

2 reviewed the usefulness of small-amplitude oscillatory shear test for determining the nature of protein matrix on samples containing protein [3]. Recognizing the importance of particle size distribution on rheological behaviour of food suspensions, this study aims to investigate the effect of particle size distribution on LVR of natural peanut butter produced from two varieties of peanut and compare their dynamic rheological behaviours with commercial peanut butter. 2.0 Materials and Methods 2. Natural peanut butter preparation Peanuts were roasted in batches of kg on two 4 x 0 cm 2 aluminium trays in a convection oven (Memmert UNB 500, Germany). The roasting conditions used were 52 C-60 minutes for the China (Virginia) and 58 C-45 minutes for the India (Spanish) peanuts. The natural peanut butter was made by grinding roasted peanut at rpm in a commercial food processor (Multi-function processor, M-PLAN Sdn. Bhd., Malaysia) for grinding time of 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, or 5.0 minutes. The peanut butter was stored in air-tight glass jar prior to analysis. A commercial smooth and spreadable peanut butter (Lady s Choice, Unilever (M) Holdings Sdn. Bhd.) was used as the reference sample. 2.2 Particle size distribution measurement The particle diameter and size distributions of peanut butter were measured using a particle size analyser, Malvern 2000 Mastersizer equipped with Hydro 2000 MU (A) sample input cell which also consists of a pump and stirrer to circulate the samples through the particle size analyzer (Malvern Instruments Ltd, Worcestershire, UK). 0.0 g of each peanut butter sample was placed in a 25 ml test tube and 5 ml of acetone was added to dilute and disperse the samples. The sample was dispersed in acetone using a vortex mixer. A transfer pipette was then used to add this diluted solution to the sample input cell of the analyzer. The sample was added until the obscuration is within the range of 0.0 to 0.5. The obscuration refers to the amount of light which is obscured by the sample because of diffraction and absorption. The acetone and the peanut butter refractive indexes were set at.376 and.500, respectively. The data were obtained and analyzed using the Malvern software for Windows (Malvern Instruments Ltd, Worcestershire, UK). Particle size calculations were based on the Mie-Scattering theory. Each sample were analysed five times and the average of them was reported. 2.3 Dynamic Rheological Measurements Peanut butter samples were subjected to dynamic rheological measurements immediately after grinding. A controlled stress rheometer (AR-G2, TA instruments, New Castle, USA), equipped with a 40mm diameter crosshatched plate-plate geometry was used to characterize the rheological behaviour of peanut butter. The stress sweep tests were performed at 25 C and constant angular frequency of Hz over a shear stress range of Pa, which covers both linear and non-linear viscoelastic range of peanut butter. The dynamic rheological data obtained included the storage modulus (G ) and the loss modulus (G ). 3.0 Results and Discussion 3. Particle size distribution Table lists the diameter on cumulative (D v, 0%,D v, 50%, D v, 90% ) of natural peanut butter as result of different grinding time. The D v, 90% is the diameter of the 90 th percentile of particle size distribution which means that 90% of the samples would have a smaller particle size than the mentioned diameter in Table. D v, 0% and D v, 50% are defined in the similar manner and represent the 0 th and 50 th percentile respectively. The diameter on cumulative (D v, 0%, D v, 50%, D v, 90% ) of commercial peanut butter is 3.72 ± 0.02, ± 0.32 and 62.7± 0.02 µm, respectively. Table : Parameters of particle size distribution of natural peanut butter samples of two peanuts varieties, Virginia from China and Spanish from India. Grinding D v,0% (µm) D v,50% (µm) D v,90% (µm) time (min) Virginia Spanish Virginia Spanish Virginia Spanish S ± ± ± ± ± ± 0.00 S ± ± ± ± ± ± 0.37

3 S ± ± ± ± ± ± 0.8 S ± ± ± ± ± ±.69 S ± ± ± ± ± ± 3.5 S ± ± ± ± ± ± 0.7 S ± ± ± ± ± ±.34 As expected, particle size distribution decreased with grinding time. The effect of grinding time was more pronounced for the D v, 90% where significant decrease can be seen with the increase of grinding time. Longer grinding time produced narrower particle size distribution for the natural peanut butter as is shown in Figure. In Figure, the particle size being represented by the calculated span, a dimensionless width factor defined as SPAN =, %, %. Between the two varieties, the India peanut was found to have a wider size distribution, % than China peanuts. Even though the size of roasted India peanuts (length: 9.22± 0.64 mm, diameter: 2.85 ± 0.5 mm) is smaller than roasted China peanuts (length: 3.9 ± 0.85 mm, diameter: 3.99 ± 0.8 mm), the results show that its structure is more resistant to breakdown during grinding. Figure : The particle size distribution indicated by span value of natural peanut butter at different grinding time. The horizontal line represents the span value of commercial peanut butter at The error bars represent the standard deviation of the mean value. Figures 2 and 3 present the mode (µm) and specific surface area (m 2 g - ) of both varieties of peanuts. Sample S 4.0 of the China variety exhibits most similarity with commercial peanut butter in terms of mode and specific surface area value. In contrast, span value of S 4.5 and S 5.0 are approximates to the commercial peanut butter (Fig. ). Generally, natural peanut butter produced from China peanuts at longer grinding time has almost equal particle size distribution with commercial peanut butter.

4 Figure 2: The mode of particle size distribution of natural peanut butter at different grinding time. The horizontal line represents the mode value of commercial peanut butter at 26.6 µm. The error bars represent the standard deviation of the mean value. Figure 3: The specific surface area (SSA) obtained from the measurement of particle size distribution of natural peanut butter at different grinding time. The horizontal line represents the SSA value of commercial peanut butter at.8 m 2 g -. The error bars represent the standard deviation of the mean value. Figures 4a and 4b show the particle size distributions of the China and India peanuts, respectively at the shortest grinding time of 2.0 and longest grinding time of 5.0 minutes. Both samples show multimodal particle size distribution with a breadth extent from 0.30 to 000 µm. Deniz et al. (2008) suggested that the presence of different and resolvable species in the sample caused multimodal histogram in the particle size distribution. Similar multimodal histograms have been reported for peanut flour [9], sesame paste [8] and chocolate as reviewed by Servais et al.(2002). Multimodal particle size distribution was found useful to increase the maximum value of volume fraction and considerably reduce the viscosity of suspensions which leads to optimum suspension transport and fluid performance during high solid content processing [4]. When multimodal distribution is obtained, the distribution is more important than its mean size thus the mean value should not be used [9]. With different size of distribution, the span value is used to differentiate the particle size of the systems. a Volume (%) commercial 2.0 minutes 5.0 minutes Particle size (µm)

5 b Volume (%) commercial 2.0 minutes 5.0 minutes Particle size (µm) Figure 4: Particle size distribution of grounded natural peanut butter for 2.0 and 5.0 minutes from (a) China and (b) India peanuts in comparison with commercial peanut butter. 3.2 Stress sweep test The comparison of the dynamic viscoelastic properties i.e. storage moduli (G ) of natural peanut butter with commercial peanut butter at 25 C are presented in Figures 5a and 5b respectively. Comparing both figures, G of the natural peanut butter from India peanuts is higher than the China suggesting association of greater elastic behavior in the earlier. Both varieties exhibits similarities in LVR. The LVRs of both varieties decreased with grinding time where in general, they are classified into two categories of grinding times, the shorter being minutes and the longer being 3.5 to 5.0 minutes, each having LVRs of 0.-0 Pa and Pa, respectively. This reflects that particle size distribution has pronounced effect on the LVR of natural peanut butter. The commercial peanut butter has a higher LVR than natural peanut butter ranging from 0.- Pa. It is because it has stabilizers which promotes peanut butter structure stability. The findings supported the microstructure study by Aryana et al. (2000) who examined the microstructure of commercial and natural peanut butter using light microscopy [5]. They discovered a well distributed protein bodies and cell wall fragments in commercial peanut butter while in natural peanut butter, there were large compact clumps, consequences of the aggregration of protein bodies and cell wall fragments. Generally, stable dispersions have longer linear regions while coagulated and strongly flocculated dispersions have relatively short linear regions [6]. In gels, the length of LVR denotes gel properties where strong gels remain in the LVR over greater stress than weak gels [7] a Storage modulus (G') S2.0 S2.5 S3.0 S3.5 S4.0 S4.5 S5.0 commercial Stress (Pa)

6 b Storage modulus (G') S2.0 S2.5 S3.0 S3.5 S4.0 S4.5 S5.0 commercial Stress (Pa) Figure 5: Stress sweeps for natural peanut butter produced from (a) China and (b) India peanuts, having different particle size distribution. The G of commercial peanut butter is also presented for comparison with the natural peanut butter. The test was conducted at 25 C, at Hz. For comparisons with other food, the LVRs of tomato pastes and ketchups occured in a range of stresses between Pa and Pa respectively [8], and the lower and high pulp content of apple puree are between 0.0 to 2 Pa and 0. to 3 Pa, respectively [] while the fat chocolate milk is 0.-9 Pa [9]. 4.0 Conclusions All peanut butter samples regardless of particle size distribution, variety and type exhibited elastic or solid-like properties from the measured G. This dynamic modulus was also greater for India peanuts than China peanuts at all respective grinding time. The LVR was greater for wider particle size distribution or shorter grinding times (S S 3.0 ) than a narrower particle size distribution from a longer grinding time (S 3.5 -S 5.0 ). It is concluded that a wider particle size distribution can produce a more stable structure of natural peanut butter. References. Hlavacek, R.G., and Freeman, W.G., Low-heat, 200 mesh milling provides smooth, uniform product. Food Processing Marketing, (4): p Citerne, P.G., Carreau, P.J. and Moan, M., Rheological properties of peanut butter.rheologicaacta, (): p Sihsobhon, C., Chompreeda, P., Suwonsichon, T., Haruthaithanasant, V. and Resurrreccion, A., American Consumer Acceptance of Satay Sauce as Affected by Different Peanut Grinding Methods, the Multi-step and One-step Processes, and Processing Times. Kasetsart J. (Nat. Sci.), : p Sato, A.C.K., and Cunha, Rosiane L., Effect of particle size on rheological properties of jaboticaba pulp. Journal of Food Engineering, (4): p Eriksson, R., Pajari, H. and Rosenholm, J.B., Shear modulus of colloidal suspensions: Comparing experiments with theory. Journal of Colloid and Interface Science, (): p Genovese, D.B., Lozano, J.E., Rao, M. Anandha, The rheology of colloidal and noncolloidal dispersions. Journal of Food Science, (2): p. R-R Rao, M.A., Measurement of viscoelastic properties of fluid and semisolid fluid, in Viscoelastic Properties of Food, M.A. Rao and J.F. Steffe, Editors. 992, Elsevier Science Publishers Ltd.: Essex, England. p Deniz,., Kahyaoglu, T., Kapucu, S. and Kaya, S., Colloidal stability and rheological properties of sesame paste. Journal of Food Engineering, (3): p Lima, L.M., Guraya, H.S. and Champagne, E.T., Improved peanut flour for a reduced-fat peanut butter product. Journal of Food Science, (5): p Subramaniam, R., and Gunasekaran, S, Small amplitude oscillatory shear studies on mozzarella cheese - part I. region of linear viscoelasticity. Journal of Texture Studies, (997): p

7 . Espinosa, L., To, N., Symoneaux, R., Renard, C.M.G.C., Biau, N. and Cuvelier, G., Effect of processing on rheological, structural and sensory properties of apple puree. Procedia Food Science, 20. (20): p Khounvilay, K., and Sittikijyothin, W., Rheological behaviour of tamarind seed gum in aqueous solutions. Food Hydrocolloids, (2): p Tunick, M.H., Small-Strain Dynamic Rheology of Food Protein Networks. Journal of Agricultural and Food Chemistry, (5): p Servais, C., Jones, R., Roberts, I., The influence of particle size distribution on the processing of food. Journal of Food Engineering, (2002): p Aryana, K.J., Resurrecion, A.V.A., Chinnan, M.S. and Beuchat, L.R., Microstructure of peanut butter stabilized with palm oil. Journal of Food Processing and Preservation, (2000): p Tabilo-Munizaga, G., and Barbosa-Canovas, G. V., Rheology for the food industry. Journal of Food Engineering, (-2): p Steffe, J.F., Rheological methods in food process engineering. 2nd. ed. Second ed. 996, Michigan: Freeman Press. 8. Bayod, E., Willers, E.P. and Tornberg, E., Rheological and structural characterization of tomato paste and its influence on the quality of ketchup. LWT - Food Science and Technology, (2008): p De Graef, V., Depypere, F., Minnaert, M. and Dewettinck, K., Chocolate yield stress as measured by oscillatory rheology. Food Research International, (9): p