Fat crystallisation behaviours in dark chocolates from varying particle size distribution (PSD) (25, 35 and 50 over-temper and under-temper), and their effects on mechanical properties and appearance evaluated.Microstructures of derived products were determined using stereoscopic binocular microscopy. Wide variations in mechanical properties and appearance were noted in products from different particle size and temper regimes. Particle size (PS) was inversely related with texture and colour, with the greatest effects noted in hardness, stickiness and lightness at all temper regimes. Over-tempering caused significant increases in product hardness, stickiness with reduced gloss and darkening of product surfaces.Under-tempering induced fat bloom in products with consequential quality defects on texture, colour and surface gloss. Micrographs revealed variations in surface and internal crystal network structure and inter-particle interactions among tempered, over-tempered and under-tempered (bloomed) samples.Under-tempering caused whitening of both surface and internal periphery of products with effects on texture and appearance. Thus, attainment of optimal temper regime during pre-crystallisation of dark chocolate was central to the desired texture and appearance as both over-tempering and under-tempering resulted in quality defects affecting mechanical properties and appearance of products.
1. Introduction
Tempering is a technique of controlled pre-crystallisation employed to induce the most stable solid form of cocoa butter, a polymorphic fat in finished chocolates. The process consists of shearing chocolate mass at controlled temperatures to promote crystallization of triacylglycerols (TAGs) in cocoa butter to effect good setting characteristics, foam stability, demoulding properties, product snap, contraction, gloss and shelf-life characteristics. Time–temperature protocols and shearing are employed to induce nucleation of stable polymorphs with the formation of three-dimensional crystal network structure influencing the microstructure, mechanical properties and appearance of products. The crystal network organisation and the polymorphic state of the TAGs crystals as affected by the crystallisation conditions are major factors determining rheological and textural properties of crystallised TAGS systems (Herrera and Hartel, 2000; Narine and Marangoni, 1999;).Lipp and Anklam, 1998; Segall). These occur as symmetric triacyglycerolsChevalley,; Beckett, 2000; Afoakwa et al., 2007b; Do et al., 2007; Afoakwa). Smaller particles improve sensory propertiesZiegler et al., 2001) but plastic viscosity and yield stress increasedistribution and ultimately sensory character in consumption and product identification.
Instrumental measurements can act as complements for sensory evaluations (Lawless and Heymann, 1998) with statistically significant correlations (Mohamed et al., 1982; Meullenet et al., 1997; Rosenthal, 1999; Ali et al., 2001; Bourne, 2002). Appropriate strategies can objectively assess features of texture and appearance such as gloss, colour, shape, roughness, surface texture, shininess, and translucency (Leemans et al., 1998; Jahns et al., 2001; Hatcher et al., 2004; Briones and Aguilera, 2005; Briones et al., 2006; Altimiras et al., 2007; Afoakwa et al., 2008a). Knowledge of tempering effects on product texture and appearance attributes can have significant commercial implications.
With recent innovations and growth in chocolate confectionery industry, understanding the factors influencing chocolate microstructure, texture and appearance would be of value in predicting changes in quality. This study was therefore aimed at investigating effects of tempering and fat crystallizations behaviors on microstructure, mechanical properties and appearance in dark chocolates varying in particle size distribution.
2. Materials and methods
2.1. Materials
Cocoa liquor of Central West African Origin was obtained from Cargill Cocoa Processing Company (York, UK); sucrose (pure cane extra fine granulated) from British Sugar Company (Peterborough, UK); pure prime pressed cocoa butter and soy lecithin from ADM Cocoa Limited (Koog aan de Zaan, Netherlands) and Unitechem Company Ltd. (Tianjin, China),respectively. The recipe, formulation and production of samples have been described previously (Afoakwa et al., 2007b). Chocolates were formulated with total fat of 35% (w/w) from sucrose, cocoa liquor, cocoa butter and lecithin. Experimental samples (5 kg batch for each formulation) were produced by mixing sucrose (40.8%) and cocoa liquor (53.7%) in a Crypto Peerless Mixer (Model K175, Crypto Peerless Ltd, Birmingham, UK) at low speed for 2 min and then at high for 3 min, then using a 3-roll refiner (Model SDX 600, Buhler Ltd., CH-9240 Uzwil, Switzerland) to a specified particle size (D90:18 ± 1 lm, 25 ± 1 lm, 35 ± 1 lm and 50 ± 1 lm) conducting particle size analysis, during refining, to ensure D90 values. The refined chocolates were melted at 50–55 _C for 24 h and the chocolate mass conched in a Lipp Conche (Model IMC-E10, Boveristr 40-42, D-68309, Mannhein, Germany) at low speed for 3.5 h at 60 _C. Lecithin (0.5%) and cocoa butter (5%) were added and then conched at high speed for 30 min to effect adequate mixing and liquefaction. Samples were kept in sealed plastic containers at ambient (20–22 _C) and moisture and fat contents determined using Karl Fischer and Soxhlet methods (ICA, 1988) and (ICA, 1990).
2.2. Determination of particle size distribution
A MasterSizer_ Laser Diffraction Particle Size Analyzer equipped with MS 15 Sample Presentation Unit (Refractive index 1.590) (Malvern Instrument Ltd., Malvern, England) was used. About 0.2 g of refined dark chocolate was dispersed in vegetable oil (Refractive index 1.450) at ambient temperature (20 ± 2 _C) until an obscuration of 0.2 was obtained. The sample was placed under ultrasonic dispersion for 2 min to ensure particles were independently dispersed and thereafter maintained by stirring during the measurement. Size distribution was quantified as the relative volume of particles in size bands presented as size distribution curves (Malvern MasterSizer_ Micro Software v 2.19). PSD parameters obtained included specific surface area, largest particle size (D90), mean particle volume (D50), smallest particle size (D10) and Sauter mean diameter (D[3,2]).
2.3. Tempering experiment
Samples were incubated at 50 _C for 4 h for melting and tempered using Aasted Mikrovert laboratory continuous three-stage tempering unit (Model AMK 10, Aasted Mikroverk A/S, Farum, Denmark). Chocolate was pumped through the multi-stage units and a worm screw drove the product through the heat exchangers.
Sensors located at specific points in the equipment measured the temperature of both the chocolate and the coolant fluid at each stage. Based on our earlier work modelling temperature controls to study tempering behaviour (Afoakwa et al., 2008b), the temperature of each of the coolant fluids (Zones 1:2:3) were thus set as 26:24:32 _C, 21:19:32 _C and 18:16:32 _C, respectively for attaining the under-tempered, optimally-tempered and over-tempered regimes. The degree of pre-crystallisation was measured using a computerized tempermeter (Exotherm 7400, Systech Analytics, Neuchâtel, Switzerland) and a built-in algorithm provided the tempering curves and temper readings in chocolate temper index (slope), corresponding to optimal temper (slope 0), undertemper (slope 1.0) and over-temper regimes (slope _1.0). The principle of this method has been described by Nelson (1999).
Chocolate from the three regimes were moulded using plastic moulds: 80 mm length; 20 mm breadth; and 8 mm height. The final products were allowed to cool in a refrigerator (5 _C) for 2 h before de-moulding onto plastic trays and conditioned at 20 ± 2 _C for 14 days before analysis. Triplicate measurements were taken for each product composition and the mean values recorded.
2.4. Texture measurements
Mechanical properties of chocolates (hardness and stickiness) were measured using TA-HD Plus Texture Analyzer with a penetration probe (needle P/2) attached to an extension bar and a 50 kg load cell and a platform reported by Afoakwa et al. (2008a). Maximum penetration and withdrawal forces through a sample (80 _ 20 mm, depth 8 mm) were determined with 8 replications at a pre-speed of 1.0 mm/s, test of 2.0 mm/s, post speed of 10.0 mm/s, penetrating 5 mm at 20 _C, converting mean values of the penetration force exerted by the 50 kg load cell into hardness (g force) and the withdrawal force with time into stickiness (g force s) data, respectively using XT.RA Dimension, Exponent 32 software (Stable Micro Systems, Godalming, Surrey, UK).
2.5. Colour and gloss measurements
HunterLab MiniscanTM XE Colorimeter Model 45/0 LAV (Hunter Associates Inc., Reston, VA) calibrated with white ceramic reference standard was used. Colour images of chocolate surfaces were converted into XYZ tristimulus values, which were further converted to CIELAB system: L*, luminance ranging from 0 (black) to 100 (white); and a* (green to red) and b* (blue to yellow) with values from _120 to +120. Information was obtained using a software algorithm (Matlab v. 6.5; The Math-Works, Inc., Natick, MA): hue angle (h_) = arctan (b*/a*); chroma (C*) = [(a*)2 + (b*)2]½. Mean values from five replicate measurements and standard deviations were calculated. Gloss of chocolate surface was measured using the multiple angle Tricor Gloss meter (805A/806H Gloss System, Elgin, IL). Reflectance was measured at an incidence light angle of 85_ from the normal to the chocolate surface, in accordance with ASTM method D523. A polished black glass plate with a refractive index of 1.567 was used as standard surface (ASTM, 1995) and given a gloss value of 200. Gloss was reported as gloss units (GU) based on determinations (in triplicate) at six positions along a chocolate sample. As a reference, a surface with a gloss value less than 10 GU is considered a low gloss surface (BYK, 1997; Briones et al., 2006).
2.6. Image acquisition and capture
A colour digital camera (Canon Powershot, Model A70, MA, USA) was mounted on a stand inside a large box with internal black surface impervious to light. Images of the optimally tempered, under- tempered and over-tempered samples were captured before storage and after 14 days in storage. The iris was operated in manual mode, with the lens aperture at f = 8 and speed 1/20 (no zoom, no flash) to achieve high uniformity and repeatability. The camera was gray balanced before each imaging session. Uniform diffuse lighting was used to illuminate the samples. The lighting system consisted of four CIE source D65 lamps (60 cm length and 18 W; Model TLD/965, Philips, Singapore) placed above the sample at a 45_ angle to maximize diffuse reflection responsible for colour. The angle between the camera lens axis and the sample was around 90_ to reduce gloss. A Kodak gray card with 18% reflectance was used as a white reference to standardize the illumination level. The gray-level image (1600 _ 1200 pixels) of this card was divided into 192 blocks, each one of 100 _ 100 pixels. After calibration, samples were placed in the field of view of the camera and an image of 1600 _ 1200 pixels (approximately covering the whole area of the tablet) was acquired and stored in JPEG (joint photographic experts group, a standard for compressing digital photographic images) format of high resolution and superfine quality.
2.7. Microstructural determinations
Chocolate samples were characterised using stereoscopic binocular microscope (Nikon, SMZ-2T, Tokyo, Japan) equipped with a variable removable lens. Micrographs (coloured images) were captured using a digital camera (Model 2.1 Rev 1, Polaroid Corporation, NY, USA) and observed using Adobe Photoshop (Version CS2, Adobe Systems Inc. NJ, USA). Triplicate experiments were conducted capturing 6 images per sample, and micrographs representing the surface of each temper regime captured and presented.Samples were then sectioned (cut) into two pieces using a knife and the internal microstructures observed.
2.8. Experimental design and statistical analysis
Two experimental variables comprising temper regime and PSD were used. Other variables including refiner temperature and pressure, conching time and temperature were held constant. A 3 _ 4 factorial experimental design was used comprising:
(i) Temper regime: optimal temper, under-temper and overtemper. (ii) PSD (D90): 18, 25, 35 and 50 lm. Statgraphics Plus 4.1 (Graphics Software System, STCC, Inc,Rockville, USA) examined mechanical properties (hardness and stiffness) and appearance (colour [L, C*, h_] and gloss) using twoway analysis of variance (ANOVA) and multiple comparison tests to determine effects of factors and their interactions. Tukey multiple comparisons (95% significance level) determined differences between levels. All experiments were conducted in triplicates and the mean values reported.
3. Results and discussion
3.1. Particle size distribution of dark chocolates
These findings (Fig. 1), previously reported (Afoakwa et al., 2008a), show volume histograms consisting of narrow (18 lm PS) and wide (25 lmPS) bimodal and narrow (35 lmPS), and wide (50 lm PS) multimodal size distributions. This PSD range 18– 50 lm using D90 values (>90% finer) covers optimum minimum and maximum sizes with direct effects on texture and sensory character in manufacture (Ziegler and Hogg, 1999; Beckett, 2000). Data from the PSD as previously described (Afoakwa et al.,2008a) showed variations in specific surface area, mean particlevolume D(v,50), Sauter mean (D[3,2]) and mean particle diameter (D[4,3]) with increasing D90 particle sizes. Specific surface area (SSA) was inversely correlated with the different component of PSD. Similar inverse relationships of SSA with all the other components of PSD have been reported (Beckett, 1999; Ziegler and Hogg, 1999; Sokmen and Gunes, 2006). Beckett (1999) concluded largest particle size and solids specific surface area are the two key parameters for chocolate manufacture. The former determines chocolate coarseness and textural character, the latter with desirable flow properties. Fat contents of the products were 35 ± 1% and moisture within the range of 0.90–0.98%.
3.2. Fat crystallisation behaviours during tempering of dark chocolate
Four different temper regimes (untempering, under-tempering, over-tempering and optimal tempering) were characterised (Fig. 2) each with its unique characteristic crystallisation behaviour. In optimal tempering, the temperature of the chocolate remained constant for sometime during cooling, to initiate formation of stable fat crystals. The crystallisation heat released was then balanced by an equal amount of cooling energy causing the growth of stable crystal nuclei in adequate amounts, which during post-tempering conditioning mature to effect shelf stability of the product. The temperature of the chocolate dropped further when the liquid cocoa butter was transformed into solid crystals resulting in solidification of the products (Fig. 2). Beckett (2000) reported that properly tempered chocolate shows formation of Form V, the most desirable polymorphic form which confers appropriate product snap, contraction, gloss and shelf-life characteristics.
Under-tempering (insufficient tempering) was caused by the relatively higher temperatures released between the multi-stage heat exchangers during tempering. The process caused development of more crystallisation heat within the product during solidification, effecting quick cooling, as more liquid fat was transformed quickly into solid form, resulting in the formation of very few stable fat crystal nuclei (Fig. 2). Distinct increase in temperature was observed at the beginning of the crystallisation, which declined again after reaching a maximum point where most of the stable crystals formed were re-melted prior to cooling. Untempered chocolate, produced no stable fat crystals as the heat exchange system generated higher crystallization heat during cooling, resulting in quick cooling of the completely melted product with no inflexion point for stable fat crystal formation (Fig. 2). Beckett (2000) explained that the crystallisation processes in both untempered and under-tempered chocolates lead to the formation of unstable Form IV polymorph, which later transforms into more stable Form VI polymorph during storage. Preliminary studies showed that untempering and under-tempering regimes exhibits different crystallisation behaviours but results in similar unstable fat crystal nucleation and growth, with similar associated storage polymorphic transformations and defects in products. Storage of the under-tempered products under ambient temperature (20–22 _C) for 14 days of conditioning induced blooming in samples, effecting various quality changes in the products as reported in this study. Products from under-tempering regime were used in this study.
Over-tempering occurred when relatively lower temperatures were exchanged between the multi-stage heat exchangers of the tempering equipment, causing significant part of the liquid fat to withdraw from the continuous phase of the chocolate, and transformed
into solid form when less liquid fat was available for pumping the product. The process released little crystallization heat during cooling, rendering a rather flat and slow cooling curve (Fig. 2). This crystallisation process results in too many small stable seed crystal formation leading to reduced strengths in the polymorphic stabilities of the fat crystals formed during the process (Talbot, 1999). As a substantial part of the phase transition (from liquid to solid) took place before the chocolate reached the mould, less contraction occurred in the mould, leading to demoulding problems with defects in final product quality and storage characters (Hartel, 2001; Lonchampt and Hartel, 2004).
3.3. Effect of temper regime and PSD on mechanical properties
Hardness showed an inverse relationship with particle sizes, with significant reductions at all temper regimes, and greatest in the under-tempered (bloomed) products (Fig. 3). Hardness of the optimally-tempered products decreased from 5318 g with 18 lm PS to 4259 g at 50 lm. Similar trends in hardness were noted with the over-tempered samples, decreasing from 6064 g with 18 lmPS to 4651 g at 50 lm, and from 6533 g with 18 lm PS to 5459 g at 50 lm in the bloomed products (Fig. 3), suggesting differences in hardness with varying PS at all temper regimes. Particle sizes have been noted as an important parameter in the hardness of fat crystal networks in many confectionery products (Narine and Marangoni, 2002; Campos et al., 2002; Marangoni and Narine 2002; Pérez- Martínez et al., 2007). Earlier studies showed inverse relationships of hardness in tempered dark chocolates with particle sizes at varying fat and lecithin levels (Afoakwa et al., 2008a), attributed to the relative strengths of their particle-to-particle interactions (Campos et al., 2002; Afoakwa et al., 2008c). Do et al. (2007) also reported consistent reductions in hardness (texture) of milk chocolates with increasing particle sizes.
The results showed that the under-tempered products had the greatest hardness (texture), attributable to the re-crystallisation process undergone by the fat in the under-tempered chocolates resulting in intense hardening of products. This trend in hardness was followed by the over-tempered samples with the optimal tempered products possessing relatively lesser hardness levels, suggesting over-tempering of dark chocolates leads to increased hardness of samples at all PS as compared to their respective optimally-tempered products.
Chocolate stickiness showed an inverse relationship with particle sizes at all temper regimes, and the greatest trends were noted in the over-tempered products (Fig. 4). Stickiness of the optimallytempered products decreased consistently from 380.67 g with 18 lm PS to 325.25 g at 50 lm. Likewise, the levels of stickiness in the over-tempered samples decreased from 447.92 g with 18 lm PS to 365.10 g at 50 lm, and from 336.86 g with 18 lm PS to 309.20 g at 50 lm in the bloomed products (Fig. 4), explaining that the over-tempered products had the greatest stickiness levels, followed by the optimally tempered products with the bloomed samples having the least. Narine and Marangoni (2001) noted that stickiness of confectionery gives information about deformability related to oral sensory characters. Analysis of variance (ANOVA) suggested significant differences (P < href="http://img03.blogcu.com/images/a/c/a/acarserpil/3_1243391246.jpg">
3.4. Effect of temper regime and PSD on colour and gloss
Lightness (L*), chroma (C*) and hue (h_) followed similar trends with varying PS at all temper regimes (Table 2). Significant (P <>
As well, the blooming caused great reductions in C* and h_ in the under-tempered products at all PS (Table 2). Hutchings (1994) stated that L*, C* and h_, respectively represent food diffuse reflectance of light, degree of saturation and hue luminance, which are dependent on particulate distribution, absorptivity and scatter-ing factors or coefficients. In a densely packed medium, scattering factor is inversely related to particle diameter (Saguy and Graf, 1991). Chocolates with varying particle sizes differ in structural and particulate arrangements influencing light scattering coefficients and thus appearance (Afoakwa et al., 2008a).
Similar decreasing trends in L* were noted in both tempered and over-tempered samples with increasing PS. However, the over-tempered samples had relatively lower L* values at all PS as compared to their corresponding optimally tempered products (Table 2).
These suggest that over-tempering reduces the degree of lightness in dark chocolates, effecting product darkening and thus affecting quality. However, no noticeable effect on C* and h_ were observed among the tempered and over-tempered products (Table 2). Thus, changes in colour in dark chocolates were primarily dependent on PS and temper regime. Bloomed dark chocolates tend to scatter more light, appear lighter and less saturated than over-tempered and optimally tempered products. The blooming process resulted in higher scattering coefficients, with subsequent paleness (whitening) - higher L* values. Hartel (1999) reported that the whitish haze in bloomed chocolate is caused by the dispersion of light of fat crystals. Similar effects of PS on the degree of whitening during blooming have been reported (Altimiras et al., 2007).Colour of foods may be affected by various optical phenomena among them scattering and surface morphology, therefore an accurate understanding of the influence of appearance on measured colour is essential.
Gloss relates to capacity of a surface to reflect directed light at the specular reflectance angle with respect to the normal surface plane (ASTM, 1995). Significant (P <>
ANOVA showed that PS and temper regime both significantly (P <>3.5. Effect of temper regime on product image
Digital images of dark chocolates (18 lm PS) were assembled to show surface appearances of optimal, under- and over-tem pered products before and after the 14 days conditioning (Fig. 5).
Initially surface appearances were similar and smooth but after 14 days, clear differences were apparent. Optimally and overtempered chocolates maintained their characteristic glossy appearance and dark brown colour but the under-tempered samples had bloomed, with appearance of surface whitish spots, rendering them dull and hazy in colour (Fig. 5). Similar increases in whiteness in under-tempered (bloomed) chocolates have been reported (Lonchampt and Hartel, 2004, 2006; Altimiras et al., 2007). Hartel (1999) explained this phenomenon as re-crystallisation of fats from a less stable Form IV to a more stable Form VI polymorph, with changes in light dispersion on small surface fat crystals (>5 lm), consequently impacting on both appearance and textural attributes. Fat bloom development, mechanisms and effects on chocolate appearance, quality and marketability has been extensively studied (Bricknell and Hartel, 1998; Ali et al., 2001; Hartel, 2001; Timms, 2003; Walter and Cornillon, 2001, 2002; Lonchampt and Hartel, 2004, 2006; Altimiras et al., 2007; Smith et al., 2007).
3.6. Effect of temper regime on microstructure
Microstructural examination using stereoscopic binocular microscopy after the 14 days conditioning showed clear variations in both surface and internal peripheries of products from varying temper regimes (Fig. 6). Over-tempered products had relatively darker surfaces and internal appearances than optimally tempered confirming the reported relative differences in L* (Table 2). Undertempered products showed both bloomed surface and internal periphery with large whitish, and distinct smaller brown spots (Fig. 6). The observed whitish appearance on surfaces and internal peripheries appear to be mixtures of re-crystallised fat and sugar crystals, and the small brown spots, cocoa solids. Lonchampt and Hartel (2004, 2006) suggested these whitish spots were primarily sugar crystals and cocoa powder and nearly devoid of fat. This difference in interpretation is the subject of further studies.
4. Conclusion
Fat crystallisation behaviour during tempering of dark chocolate play vital roles in defining the structure, mechanical properties and appearance of products. Wide variations in mechanical properties and appearance occurred in products from different PS and temper regimes. Particle size was inversely related with texture and colour, with the greatest effects noted with hardness, stickiness and lightness at all temper regimes. Over-tempering caused increases in product hardness, stickiness with reduced gloss and darkening of product surfaces. Under-tempering induced fat bloom in products with consequential quality defects in texture, colour and surface gloss. Micrographs revealed clear variations in surface and internal crystal network structure and inter-particle interactions among tempered, over-tempered and under-tempered (bloomed) samples. Blooming caused whitening of both surface and internal periphery of products with consequential effects on texture and appearance. Hence, attainment of optimal temper during tempering (pre-crystallisation) of dark chocolate is vital to the desired texture and appearance of products, as both over-tempering and under-tempering result in quality defects affecting mechanical properties and appearance of products.
Acknowledgements
This study was co-funded by the Government of Ghana and Nestlé Product Technology Centre (York, UK). The sponsors are gratefully acknowledged for the Research Support. We also wish to thank Drs. Steve Beckett, Angela Ryan, John Rasburn and Angel Manez (Nestlé PTC, York) for useful technical discussions.
