- Email: [email protected]
Effect of radio frequency (RF) heating on the texture, colour and sensory properties of a comminuted pork meat product
Food Research International 38 (2005) 337–344 www.elsevier.com/locate/foodres
Effect of radio frequency (RF) heating on the texture, colour and sensory properties of a comminuted pork meat product Nigel P. Brunton, James G. Lyng *, Wenqu Li, Denis A. Cronin, Desmond Morgan, Brian McKenna Department of Food Science, Faculty of Agriculture, University College Dublin, Belfield, Dublin 4, Ireland Received 14 January 2004; accepted 7 June 2004
Abstract Radio frequency (RF) cooking is a form of dielectric heating in which products are heated by subjecting them to an alternating electromagnetic field between two parallel electrodes. Although similar in some respects to Microwave heating, RF has been proposed to be more suitable for industrial heating of meats because of the greater penetration depths possible with this technology. In this study an RF cooking protocol was developed and its effect on selected quality attributes of pork based white pudding was examined. Whilst cooking of the product in air proved unfeasible due to arcing, use of a polyethylene cell with circulating hot water (80 C) facilitated successful heating of the product. Application of RF using an optimised cooking protocol (RF power = 450 W, cell volume = 500 ml and continuous circulation) resulted in a mean end-point temperature of 73 C after 7 min 40 s. Similar mean end-point temperatures in water bath and steam oven heated products were achieved after 29 and 33 min, respectively. A factorial experiment was conducted to assess selected quality attributes of the cooked puddings. Results show that RF heated puddings were not significantly different (P > 0.05) from water bath and steam oven heated products with regard to instrumental colour, instrumental texture (Kramer shear and texture profile analysis) and expressible fluid. Furthermore, results of a sensory similarity test involving 60 panellists indicated that panellists were not able to detect differences between puddings cooked by RF and conventional methods. Overall this suggests that RF heating technology could have potential in pasteurisation of meat products though further work is needed to verify this. 2004 Elsevier Ltd. All rights reserved. Keywords: Radio frequency; Comminuted meat; Texture; Colour; Sensory
1. Introduction Traditionally, pasteurisation of meat products has been carried out by immersion in hot water or by steam cooking. In both these processes heat is transferred by conduction from the outside surface of the product to its interior. This can lead to over-heating of the outer regions of the product while waiting for the interior to reach appropriate temperatures, which in turn can *
Corresponding author. Tel.: +353 1 716 7710; fax: +353 1 716 1147. E-mail address: [email protected] (J.G. Lyng). 0963-9969/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodres.2004.06.016
potentially reduce product quality and nutritional value. In contrast radio frequency (RF) heating occurs volumetrically which essentially means that all parts of the product more or less heat simultaneously and at the same rate. This largely avoids overheating in the outer regions of the product and could potentially increase product quality and nutritional value while reducing cooking times. Previous investigations have evaluated the potential of this technology for the pasteurisation of uncased sausage meat emulsions while they were continuously pumped through a low loss plastic tube (Houben, Schoenmakers, van Putten, van Roon, & Krol, 1991). Furthermore, in the 1990s a joint venture between
338
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
Tulip International AS Denmark and APV, attempted to apply this technology on a commercial basis to uncased meats while they were again pumped through a plastic tube. While these attempts were widely acclaimed and received a number of innovation awards, this particular method of application is no longer practiced. Whilst RF heating has been touted as having potential economic benefits for the heat processing of foodstuffs and the technique has been applied to the post-baking drying of biscuits (Anon., 1985, 1995) and the thawing of meats (Bengtsson, 1963; Mc Cormick, 1988) limited information is available on the eating quality of RF produced foodstuffs or how they compare to products produced by conventional methods (Van Roon, Houben, Koolmees, & van Vilet, 1994). While cooking meat continuously as it is pumped through a tube is no longer practiced, RF technology could also be applied to cased meats if a suitable application method was developed. This could reduce capital costs substantially as the products could be cooled while still encased, which would fit in more with conventional cooked meat production lines and practices. The objective of this work was to develop a system for the production of encased cooked meats using RF heating. Another important objective was the comparison of the quality of the resultant RF heated products with those produced by commercial cooking methods.
2. Materials and methods 2.1. Meat handling Lean pork shoulder and pork back fat was obtained from a local producer (Galtee meats, Cork, Ireland) and ground through a 10 mm plate using a mechanical mincer (Model TS8E, Tritacarne, Omas, Italy). The ground tissue was then placed in polyethylene bags, vacuum packaged using a Webomatic vacuum packaging system (Model No. 021ODC681, Webomatic, Bochum, Germany) and stored at 18 C until required for product manufacture. 2.2. WP manufacture White pudding (WP) was manufactured in 12.2 kg batches using the following ingredients Lean pork (Galtee meats, Cork, Ireland) (5.3 kg), pork fat (Galtee meats) (2.5 kg), kibbled onion (National Food Ingredients, Limerick, Ireland) (0.3 kg), seasoning (National Food Ingredients) (0.3 kg), iced water (2.4 kg), rusk meal (William Blake Ltd., Dublin, Ireland) (1.2 kg) and cure solution (0.2 kg). The cure solution consisted of water (81%), salt (12.5%), sodium tri polyphosphate (2.0%), sodium ascorbate (0.25%) and sodium nitrite (0.1%). Appropriate quantities of the frozen lean and
fat were air-tempered at 5 C for 24 h prior to each batter preparation. Processing of the batters involved blending the thawed minced meat and fat, cure solution, seasoning and water in a Manica bowl chopper (Model No. CM22, Equipaimentos Carnicos, Barcelona, Spain) at knife and bowl speed 1 for 120 s. Following this the onion and rusk meal were added and the mixture chopped for an additional 120 s at knife and bowl speed 2. After blending the WP batter was filled into casings (Walsrode K-Plus, Casetech, GMBH, Walsrode, Germany) using a mechanical filler (Model No. EM-12, Equipaimentos Carnicos) to a weight of 0.2 kg (±0.0005 kg) and sealed with plastic cable-ties (Maplin Electronics, Dublin, Ireland). Three, 12.2 kg batches of product were prepared. The puddings prepared from each batch were then randomly divided into four groups and each group was assigned to one of the cooking protocols (outlined in Sections 2.3 and 2.4) following which they were stored at 20 C until required. 2.3. RF cooking of WPÕs Previous work on RF heating of cased meat products in air proved unsatisfactory due to arcing and burning of the product and casing. Experiments conducted suggested that immersion of the product in water avoided this phenomenon. All RF heated products were processed in a customised RF batch oven (Capenhurst Technologies Ltd., Capenhurst Technology Park, Capenhurst, Chester CH1 6ES, England), which was set at 450 W in this work but which had a maximum power output of 600 W. For RF cooking, samples were placed in a purposed built circulating water cell. The material used was semi crystalline 60 mm thick high density polyethylene (Alperton Engineering Ltd., Dublin Industrial Estate, Glasnevin, Dublin, Ireland). The basic cell design was based on that outlined by Bengtsson and Green (1970) and therefore was fitted with secondary electrodes. However, substantial modifications were made to allow for the circulation of hot water (80 C) around the product at all times. This was achieved using a circulating water bath (WB) (Model LTD 20, Grant Instruments, Cambridge, UK) and had the added benefit of allowing a holding period after RF heating. The cell was also fitted with a side mounted screw type lid (which enabled product to be introduced through the side of the cell) and incorporated a port to allow for the introduction of fibre optic probes (for monitoring temperature during cooking). Sealing of the cell was achieved through the use of screws and a countersunk rubber gasket. Fig. 1(a) and (b) provide graphical and diagrammatic illustrations of the structure of this cell. This cell was placed on the lower electrode of the RF oven and an air gap was left between it and the upper electrode.
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
339
Fig. 1. (a) Schematic of radio frequency (RF) heating cell and its components. (b) Circulating hot water heating cell in radio frequency oven.
2.4. Conventional cooking of WPÕs For steam processing samples were cooked in a thermostatically controlled KERRES smoke-air steam oven (Type CS 350, Raicher-und-Kochanlagen, D-71560 Sulzbach-Murr, Germany) set at 80 C.
The other conventional method used was hot water immersion. WPs were cooked in thermostatically controlled hot waterbath (WB) (Model No. Y38, Grant Instruments (Cambridge) Ltd., Shepreth, Royston, Herts SG8 6PZ, England) set at 80 C.
340
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
2.5. Cooking protocol development In a preliminary study, time temperature profiles were recorded during WB, steam and RF cooking. During steam cooking, profiles were recorded using thermocouples attached to a data logger (Squirrel, Model No. 1600, Grant Instruments Ltd., Barrington, Cambridge CB2 5QZ, England). For RF cooking Luxtron fluoroptic thermometry (Model No. 790, Luxtron Corporation, 2775 North Western Parkway, Santa Clara, CA) was used to record temperatures in RF cooked samples. Conventional (steam and WB) and RF (RF1) cooking protocols were developed which were in compliance with typical commercial cooking guidelines (i.e. end point temperature in the coldest spot of 73 C for 2 min). From the time temperature data recorded during RF and Steam cooking, pasteurisation units (PU60) were calculated. For PU60 values a z value for pasteurisation (zp) value of 5.5 C (Listeria monocytogenes in meat products, Institute of Food Technologists, 2001), a reference temperature (hp) of 60 C and Eq. (1) (based on equations provided in Stumbo (1973) was used) Z tp Z tp d t 10T hp =zp ¼ d t 10T 60=5:5 0
0
¼ Pasteurisation units ðPU60 Þ,
ð1Þ
where T is the temperature (C) at the measurement point at any time during the heat process, dt is the duration of time (min) at a particular temperature and tp is the time (min) at the end of the heating process. From this data a second RF cooking protocol (RF2) was developed (by extending the holding time) which produced a comparable PU60 to the conventionally cooked samples. This in total four cooking protocols were developed: (1) steam cooked to 73 C for 2 min (33 min at 80 C in an S); (2) WB cooked to 73 C for 2 min (29 min in 80 C in a WB); (3) RF cooked to 73 C for 2 min (7 min 30 s at 450 W RF with 80 C circulating water); and (4) RF cooked to 73 C for 2 min followed by an extended holding time to produce a similar PU60 to that produced in steam cooked samples (7 min 40 s at 450 W RF with 80 C circulating water). 2.6. Instrumental texture measurement An Instron, Universal Testing Machine (Merlin, Model No. 5544, Instron Corporation, Highwycombe, England) was used for all textural measurements in conjunction with the Instron Merlin software package (Version 2019). Textural profile analysis (TPA) was conducted using the methods of Bourne (1978) as described by Colmenero, Barreto, Mota, and Carballo (1995). The Penetration test was performed using the method of Cavestany, Colmenero, Solas, and Carballo (1994). For Kramer shear analysis, slices (2.5 cm ·
5.5 cm · 4 mm) were prepared using a template. These samples were immediately wrapped in cling-film and allowed to equilibrate to room temperature. The WB analysis was performed at a crosshead speed of 200 mm min1 and both load and extension were analysed. For each of the textural analyses conducted, puddings from three separately prepared batches were assessed with two puddings taken per batch and six readings taken from each pudding. 2.7. Expressible fluid test (EFT) Preweighed meat cores (2.5 cm diameter, 2 cm high) were prepared following cooking and cooling of the meat emulsion batter. Cores were immediately wrapped in cling film and allowed to equilibrate to room temperature for 2 h. Cores were then removed from cling film, placed between three sheets of preweighed Whatman #1 filterpaper (9 cm diameter) with two sheets underneath and one above the core. Samples were compressed to 75% of their total height using an Instron (Model No. 5544, Instron High Wycombe, Bucks, UK) operating at a crosshead speed of 50 mm min1. Total expressible fluid (TEF) consisting of expressible moisture and expressible suspended solids were calculated using the methods of Lee, Whiting, and Jenkins (1987) and Cavestany et al. (1994). For each of the textural analyses conducted, puddings from three separately prepared batches were assessed with two puddings taken per batch and six readings taken from each pudding. 2.8. Colour measurement Colour measurements were performed on cores prepared for TPA, prior to TPA measurement. A chroma meter (Model CR-300 Minolta, Minolta (UK) Limited, 1-3 Tanners Drive, Blakelands North, Milton Keynes, MK14 5BU, England) was used to determine Hunter L* (lightness), a* (redness/greenness) and b* (yellowness/blueness) from which the hue angle (H) and the saturation (S) were calculated using the equations give by MacDougall and Rhodes (1972). The chroma meter was calibrated for internal light (D65) before carrying out colour measurements. As readings were taken from TPA and penetration test samples, puddings from three separately prepared batches were assessed with four puddings taken per batch, six cores per pudding and two readings per core. 2.9. Sensory analysis A triangle taste test was conducted to evaluate whether panellists can distinguish between WB cooked samples and samples prepared by RF2. Untrained panellists were recruited to carry out the evaluation. Prior to each session samples were cut into 20 mm thick slices
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
and grilled using a domestic cooker for 500 s. Each panellist was given three samples (two identical and one different) of cooked WP and asked to record the number of the odd sample. A separate batch of puddings was prepared for sensory analysis and testing was continued until a significant result was obtained.
341
Table 1. When ANOVA was conducted on each of the attributes assessed, no significant differences between the TPA, Kramer shear or penetration test values of products cooked by the different methods was observed (P > 0.05). 3.3. Expressible fluid analysis
2.10. Statistical analysis One-way analysis of variance (ANOVA) was used to test the effect of cooking protocol on the quality evaluation assessed using the SAS statistical analysis software (Version 8.2, Statistical Analysis Systems, Cary, NC, USA). For sensory analysis a sequential test as described by Meilgaard, Civille, and Carr (1991) was used to analyse the data.
3. Results 3.1. Temperature profiles and lethality Fig. 2 shows a time–temperature profile for a cased WP cooked using a heated circulating water cell. A time temperature profile for an identical pudding heated using a steam-oven is provided for comparison. A standard cooking time of 7 min 30 s was required to achieve a temperature of 73 C while for steam cooked samples this was only achieved after 33 min. Whilst both RF and steam cooked samples achieved similar end-point temperatures, a further 10 s holding time was required for RF cooked samples receive an identical lethality to the steam cooked samples. 3.2. Instrumental texture A range of textural parameters for WP cooked using RF or conventional cooking protocols are presented in
Results of Expressible fluid analysis of WP cooked using RF or conventional cooking protocols are presented in Table 2. No significant differences in yield expressible fluid (%) and expressible moisture (%) were recorded between products cooked by the various cooking methods (P > 0.05). 3.4. Instrumental colour Results of instrumental colour analysis of WP cooked using RF or conventional cooking protocols are presented in Table 3. No significant differences in colour (i.e. hue angle, saturation, L, a* and b* values) were recorded between products cooked by the various cooking methods (P > 0.05). 3.5. Sensory analysis Results for the sequential test are plotted in Fig. 3. Equations provided by Meilgaard et al. (1991) were used to plot lines on this graph with number of panellists on the x-axis and number of correct tests on the y-axis. These lines divide the plotted area into three regions: the acceptance region (i.e. accept that samples are similar), the rejection region (reject that samples are similar), and the continuing testing region. Testing was continued until the plotted line crossed one of the lines bordering the continue testing region. In this work it was found necessary to use 57 panellists before it could be concluded that no perceivable difference exists between the samples (P > 0.05).
90
4. Discussion
80
Temperature (oC)
70 60 50
RF
40
S
30 20 10 0 0
5
10
15 20 Time (min)
25
30
35
Fig. 2. Typical time–temperature profiles of radio frequency (RF) and steam (S) cooked white pudding.
RF heating offers the advantage of a reduction in processing times over conventional methods. However, any advantage conferred on meat processors by this reduction would be lost if application of the technology resulted in a loss of product quality. In this light the lack of any significant differences (P P 0.05) in the textural and colour parameters between RF1, RF2 and conventionally processed WPÕs (Table 1) suggests this technology looks promising for this particular product. In addition, triangle tests of WP cooked by RF2 and steam cooking methods indicated that panellists could not differentiate between the conventionally and RF cooked samples. In contrast Van Roon et al. (1994) reported
342
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
Table 1 Texture of white pudding cooked by water bath (WB), steam (S) and radio frequency (RF) methods Instrumental texture attribute
RF1a
RF2b
Mean Kramer shear Load (N) Extension (mm) Texture profile analysis Chewiness (N mm) Cohesiveness Cohesion force (N mm2) Springiness (mm) Hardness1 (N) Hardness 2 (N)
b c d
Mean
SD
WBd
Mean
SD
Mean
P-value SD
86.4 7.61
11.05 3.238
84.5 7.55
16.492 3.326
86.2 8.054
13.928 3.905
89.4 8.03
12.374 3.887
0.972 0.996
194.4 0.515 77.8 7.75 48.9 37.0
25.65 0.041 1.693 0.161 3.174 3.124
177.9 0.503 76.9 7.69 45.5 34.7
41.94 0.054 2.378 0.239 7.587 8.223
223.2 0.516 79.6 7.94 54.2 41.3
53.37 0.082 1.655 0.160 8.537 9.207
203.3 0.482 79.6 7.94 53.5 38.0
31.52 0.063 0.755 0.075 4.366 6.203
0.577 0.835 0.726 0.723 0.467 0.622
2.89 5.96
0.609 0.389
2.75 6.06
0.427 0.383
2.94 5.86
0.380 0.291
2.79 5.47
0.394 0.684
0.895 0.198
Penetration test Penetration test load (N) Penetration elastic behaviour (mm) a
SD
Sc
RF1: 450 W for 7 min 30 s with 80 C circulating water followed by 2 min holding time in 80 C water. RF2: as RF1 but holding period extended by 10 s. S: cooked for 33 min at 80 C in a steam oven. WB: cooked for 29 min in water at 80 C water.
Table 2 Yield and expressible fluid of white pudding cooked by water bath (WB), steam (S) and radio frequency (RF) methods Instrumental quality parameter
Yield (%) Total expressible fluid (%) Expressible moisture (%) a b c d
RF1a
RF2b
Sc
WBd
P-value
Mean
SD
Mean
SD
Mean
SD
Mean
SD
0.999 0.019 0.833
0.004 0.004 0.03
1.000 0.018 0.837
0.001 0.006 0.028
1.000 0.019 0.832
0.001 0.005 0.028
1.000 0.018 0.807
0.001 0.004 0.044
0.342 0.980 0.698
RF1: 450 W for 7 min 30 s with 80 C circulating water followed by 2 min holding time in 80 C water. RF2: as RF1 but holding period extended by 10 s. S: cooked for 33 min at 80 C in a steam oven. WB: cooked for 29 min in water at 80 C water.
Table 3 Colour attributes of white pudding cooked by water bath (WB), steam (S) and radio frequency (RF) methods Instrumental colour attribute
L* Hue angle () Saturation a* b* a b c d
RF1a
RF2b
Sc
WBd
P-value
Mean
SD
Mean
SD
Mean
SD
Mean
SD
61.468 47.465 9.366 6.322 6.898
3.147 2.845 0.931 0.667 0.792
61.485 47.412 9.301 6.283 6.848
3.175 2.53 0.941 0.624 0.813
61.671 46.496 9.393 6.466 6.804
3.482 2.43 1.003 0.794 0.727
61.856 47.584 9.422 6.340 6.955
3.634 3.2 1.075 0.739 0.933
0.997 0.917 0.977 0.977 0.992
RF1: 450 W for 7 min 30 s with 80 C circulating water followed by 2 min holding time in 80 C water. RF2: as RF1 but holding period extended by 10 s. S: cooked for 33 min at 80 C in a steam oven. WB: cooked for 29 min in water at 80 C water.
that RF cooked meat doughs were much firmer than WB cooked samples. In addition, Laycock, Piyasena, and Mittal (2003) reported that RF cooked comminuted beef samples had significantly lower hardness 1 and hardness 2 values indicating that the RF product was more ÔsoggyÕ than the conventionally cooked product.
The products described by Laycock et al. (2003) were comminuted beef with no other added ingredients. Therefore they did not contain salt or other added binders, which would have helped them to retain water. Furthermore, they would not have been representative of typical commercial meat doughs unlike the products in
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
343
Fig. 3. Sensory analysis test plot of results of radio frequency cooked vs. steam cooked white pudding.
this work. These authors also reported differences in springiness and gumminess between RF and WB treated samples. In this work RF cooking resulted in a reduction of cooking times of up to a third when compared to steam cooking. Laycock et al. (2003) reported that a reduction in cooking time of up to 10 fold could be achieved for RF cooked comminuted beef as compared to WB cooked samples. The workers (Laycock et al., 2003) reporting the greater reduction in cooking time used a substantially higher RF input power (696 W versus 450 W in the present case) and thus a relatively greater reduction in cooking time would be expected. The RF input power in the present study was limited by the maximum power output of the RF generator (600 W). Another contrast between previously reported work and the present study was that in this work, similar to the practice used in commercial meat processing operations, the WPÕs (for both RF and conventional cooking) were sealed in plastic casings which ensured that the product maintained a uniform shape during processing. In addition, however, casings help to prevent post-process contamination and eliminate evaporation during cooling and loss of juices, thereby encouraging gel formation and moisture retention in the product. In contrast, packaging of the product used by Laycock et al. (2003) differed between conventional and RF cooking. Products for conventional cooking were packed in a plastic bag during cooking whereas the RF cooked samples were placed without casing, in a polytetraflouroethylene (PTFE) cylinder between the RF electrodes. A principal objective of the present work was the development of a RF cooking protocol that could be integrated with existing commercial production methods but which would allow cooking to take place in a continuous fashion rather than the normal batch type cooking methods which are predominantly used for these products. The authors of the present study suggest that this could be achieved by placing the product to be heated on a conveyor belt and passing it between RF electrodes
immersed in a heated WB. Once the product has been heated to a suitable end-point temperature by application of RF heating, a subsequent holding phase could be achieved by passage of the product through a heated WB without electrodes.
5. Conclusion RF cooking of products in air was found to be unfeasible and therefore it was necessary to surround products with heated water during RF cooking. A cell made from high-density polyethylene, which held the product and allowed water circulation facilitated successful heating of the product. This study has demonstrated that 450 W of RF power could reduce cooking times to 7 min 40 s as compared to 33 min for conventional cooking. Instrumental quality indices of the resultant product revealed no significant differences (P P 0.05) with regard to texture and colour. Subsequent sensory work confirmed this observation. These results suggested that RF heating could be very suitable for industrial heating of encased WP. However, similar studies would be required on a large range of meat products before this conclusion could be reached. Future work should also compare the effect of higher RF output powers and the resultant shorter cooking times on product quality.
References Anon. (1985). Radiofrequency and the production of food. Baking Today, September, 46–47. Anon. (1995). Cooking with radio frequency. Meat International, 5(5), 10–11. Bengtsson, N. E. (1963). Electronic defroisting of meat and fish at 35 and 2450 mcs-a laboratory comparision. Food Technology, (October), 97–100. Bengtsson, N. E., & Green, W. (1970). Radio frequency pasteurisation of cured hams. Journal of Food Science, 35, 681–687.
344
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
Bourne, M. C. (1978). Texture profile analysis. Food Technology, 7, 62–72. Cavestany, M., Colmenero, F. J., Solas, M. T., & Carballo, J. (1994). Incorporation of sardine surimi in bologna sausage containing different fat levels. Meat Science, 38, 27–37. Colmenero, F. J., Barreto, G., Mota, N., & Carballo, J. (1995). Influence of protein and fat content and cooking temperature on texture and sensory evaluation of bologna sausage. LebensmittelWissenschaft Und-Technologie, 28, 481–487. Houben, J., Schoenmakers, L., van Putten, E., van Roon, P., & Krol, B. (1991). Radio-frequency pasteurisation of sausage emulsions as a continuous process. Journal Microwave Power and Electronic Energy, 26(4), 202–205. Institute of Food Technologists (2001). Overarching principles: kinetics and pathogens of concern for all technologies. In D. E. Weber, F. R. Katz, & C. L. Mattson (Eds.), Journal of Food Science Special Supplement: Kinetics of microbial inactivation for alternative food processing technologies (pp. 16–32). Chicago, IL: Institute of Food Technologists.
Lee, C. M., Whiting, R. C., & Jenkins, R. K. (1987). Texture and sensory evaluations of frankfurters made with different formulations and processes. Journal of Food Science, 52, 898–900. Laycock, L., Piyasena, P., & Mittal, G. S. (2003). Radio frequency cooking of ground, comminuted and muscle meat products. Meat Science, 65, 959–965. MacDougall, D. B., & Rhodes, D. N. (1972). Characteristics of the appearance of meat III. Studies on the colour of meat from young bulls. Journal of the Science of Food and Agriculture, 23, 637–647. Mc Cormick, R. (1988). Dielectric heat seeks low moisture applications. Prepared Foods, 162(9), 139–140. Meilgaard, M., Civille, G. V., & Carr, B. T. (1991). Sensory evaluation techniques (2nd ed.). London: CRC Press. Stumbo, C. R. (1973). Thermobacteriology in food processing. London: Academic Press. Van Roon, P. S., Houben, J. H., Koolmees, P. A., & Van Vilet, T. (1994). Mechanical and microstructural characteristics of meat doughs, either heated by a continuous process in a radio-frequency field or conventionally in a waterbath. Meat Science, 38(1), 103–116.
Effect of radio frequency (RF) heating on the texture, colour and sensory properties of a comminuted pork meat product Nigel P. Brunton, James G. Lyng *, Wenqu Li, Denis A. Cronin, Desmond Morgan, Brian McKenna Department of Food Science, Faculty of Agriculture, University College Dublin, Belfield, Dublin 4, Ireland Received 14 January 2004; accepted 7 June 2004
Abstract Radio frequency (RF) cooking is a form of dielectric heating in which products are heated by subjecting them to an alternating electromagnetic field between two parallel electrodes. Although similar in some respects to Microwave heating, RF has been proposed to be more suitable for industrial heating of meats because of the greater penetration depths possible with this technology. In this study an RF cooking protocol was developed and its effect on selected quality attributes of pork based white pudding was examined. Whilst cooking of the product in air proved unfeasible due to arcing, use of a polyethylene cell with circulating hot water (80 C) facilitated successful heating of the product. Application of RF using an optimised cooking protocol (RF power = 450 W, cell volume = 500 ml and continuous circulation) resulted in a mean end-point temperature of 73 C after 7 min 40 s. Similar mean end-point temperatures in water bath and steam oven heated products were achieved after 29 and 33 min, respectively. A factorial experiment was conducted to assess selected quality attributes of the cooked puddings. Results show that RF heated puddings were not significantly different (P > 0.05) from water bath and steam oven heated products with regard to instrumental colour, instrumental texture (Kramer shear and texture profile analysis) and expressible fluid. Furthermore, results of a sensory similarity test involving 60 panellists indicated that panellists were not able to detect differences between puddings cooked by RF and conventional methods. Overall this suggests that RF heating technology could have potential in pasteurisation of meat products though further work is needed to verify this. 2004 Elsevier Ltd. All rights reserved. Keywords: Radio frequency; Comminuted meat; Texture; Colour; Sensory
1. Introduction Traditionally, pasteurisation of meat products has been carried out by immersion in hot water or by steam cooking. In both these processes heat is transferred by conduction from the outside surface of the product to its interior. This can lead to over-heating of the outer regions of the product while waiting for the interior to reach appropriate temperatures, which in turn can *
Corresponding author. Tel.: +353 1 716 7710; fax: +353 1 716 1147. E-mail address: [email protected] (J.G. Lyng). 0963-9969/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodres.2004.06.016
potentially reduce product quality and nutritional value. In contrast radio frequency (RF) heating occurs volumetrically which essentially means that all parts of the product more or less heat simultaneously and at the same rate. This largely avoids overheating in the outer regions of the product and could potentially increase product quality and nutritional value while reducing cooking times. Previous investigations have evaluated the potential of this technology for the pasteurisation of uncased sausage meat emulsions while they were continuously pumped through a low loss plastic tube (Houben, Schoenmakers, van Putten, van Roon, & Krol, 1991). Furthermore, in the 1990s a joint venture between
338
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
Tulip International AS Denmark and APV, attempted to apply this technology on a commercial basis to uncased meats while they were again pumped through a plastic tube. While these attempts were widely acclaimed and received a number of innovation awards, this particular method of application is no longer practiced. Whilst RF heating has been touted as having potential economic benefits for the heat processing of foodstuffs and the technique has been applied to the post-baking drying of biscuits (Anon., 1985, 1995) and the thawing of meats (Bengtsson, 1963; Mc Cormick, 1988) limited information is available on the eating quality of RF produced foodstuffs or how they compare to products produced by conventional methods (Van Roon, Houben, Koolmees, & van Vilet, 1994). While cooking meat continuously as it is pumped through a tube is no longer practiced, RF technology could also be applied to cased meats if a suitable application method was developed. This could reduce capital costs substantially as the products could be cooled while still encased, which would fit in more with conventional cooked meat production lines and practices. The objective of this work was to develop a system for the production of encased cooked meats using RF heating. Another important objective was the comparison of the quality of the resultant RF heated products with those produced by commercial cooking methods.
2. Materials and methods 2.1. Meat handling Lean pork shoulder and pork back fat was obtained from a local producer (Galtee meats, Cork, Ireland) and ground through a 10 mm plate using a mechanical mincer (Model TS8E, Tritacarne, Omas, Italy). The ground tissue was then placed in polyethylene bags, vacuum packaged using a Webomatic vacuum packaging system (Model No. 021ODC681, Webomatic, Bochum, Germany) and stored at 18 C until required for product manufacture. 2.2. WP manufacture White pudding (WP) was manufactured in 12.2 kg batches using the following ingredients Lean pork (Galtee meats, Cork, Ireland) (5.3 kg), pork fat (Galtee meats) (2.5 kg), kibbled onion (National Food Ingredients, Limerick, Ireland) (0.3 kg), seasoning (National Food Ingredients) (0.3 kg), iced water (2.4 kg), rusk meal (William Blake Ltd., Dublin, Ireland) (1.2 kg) and cure solution (0.2 kg). The cure solution consisted of water (81%), salt (12.5%), sodium tri polyphosphate (2.0%), sodium ascorbate (0.25%) and sodium nitrite (0.1%). Appropriate quantities of the frozen lean and
fat were air-tempered at 5 C for 24 h prior to each batter preparation. Processing of the batters involved blending the thawed minced meat and fat, cure solution, seasoning and water in a Manica bowl chopper (Model No. CM22, Equipaimentos Carnicos, Barcelona, Spain) at knife and bowl speed 1 for 120 s. Following this the onion and rusk meal were added and the mixture chopped for an additional 120 s at knife and bowl speed 2. After blending the WP batter was filled into casings (Walsrode K-Plus, Casetech, GMBH, Walsrode, Germany) using a mechanical filler (Model No. EM-12, Equipaimentos Carnicos) to a weight of 0.2 kg (±0.0005 kg) and sealed with plastic cable-ties (Maplin Electronics, Dublin, Ireland). Three, 12.2 kg batches of product were prepared. The puddings prepared from each batch were then randomly divided into four groups and each group was assigned to one of the cooking protocols (outlined in Sections 2.3 and 2.4) following which they were stored at 20 C until required. 2.3. RF cooking of WPÕs Previous work on RF heating of cased meat products in air proved unsatisfactory due to arcing and burning of the product and casing. Experiments conducted suggested that immersion of the product in water avoided this phenomenon. All RF heated products were processed in a customised RF batch oven (Capenhurst Technologies Ltd., Capenhurst Technology Park, Capenhurst, Chester CH1 6ES, England), which was set at 450 W in this work but which had a maximum power output of 600 W. For RF cooking, samples were placed in a purposed built circulating water cell. The material used was semi crystalline 60 mm thick high density polyethylene (Alperton Engineering Ltd., Dublin Industrial Estate, Glasnevin, Dublin, Ireland). The basic cell design was based on that outlined by Bengtsson and Green (1970) and therefore was fitted with secondary electrodes. However, substantial modifications were made to allow for the circulation of hot water (80 C) around the product at all times. This was achieved using a circulating water bath (WB) (Model LTD 20, Grant Instruments, Cambridge, UK) and had the added benefit of allowing a holding period after RF heating. The cell was also fitted with a side mounted screw type lid (which enabled product to be introduced through the side of the cell) and incorporated a port to allow for the introduction of fibre optic probes (for monitoring temperature during cooking). Sealing of the cell was achieved through the use of screws and a countersunk rubber gasket. Fig. 1(a) and (b) provide graphical and diagrammatic illustrations of the structure of this cell. This cell was placed on the lower electrode of the RF oven and an air gap was left between it and the upper electrode.
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
339
Fig. 1. (a) Schematic of radio frequency (RF) heating cell and its components. (b) Circulating hot water heating cell in radio frequency oven.
2.4. Conventional cooking of WPÕs For steam processing samples were cooked in a thermostatically controlled KERRES smoke-air steam oven (Type CS 350, Raicher-und-Kochanlagen, D-71560 Sulzbach-Murr, Germany) set at 80 C.
The other conventional method used was hot water immersion. WPs were cooked in thermostatically controlled hot waterbath (WB) (Model No. Y38, Grant Instruments (Cambridge) Ltd., Shepreth, Royston, Herts SG8 6PZ, England) set at 80 C.
340
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
2.5. Cooking protocol development In a preliminary study, time temperature profiles were recorded during WB, steam and RF cooking. During steam cooking, profiles were recorded using thermocouples attached to a data logger (Squirrel, Model No. 1600, Grant Instruments Ltd., Barrington, Cambridge CB2 5QZ, England). For RF cooking Luxtron fluoroptic thermometry (Model No. 790, Luxtron Corporation, 2775 North Western Parkway, Santa Clara, CA) was used to record temperatures in RF cooked samples. Conventional (steam and WB) and RF (RF1) cooking protocols were developed which were in compliance with typical commercial cooking guidelines (i.e. end point temperature in the coldest spot of 73 C for 2 min). From the time temperature data recorded during RF and Steam cooking, pasteurisation units (PU60) were calculated. For PU60 values a z value for pasteurisation (zp) value of 5.5 C (Listeria monocytogenes in meat products, Institute of Food Technologists, 2001), a reference temperature (hp) of 60 C and Eq. (1) (based on equations provided in Stumbo (1973) was used) Z tp Z tp d t 10T hp =zp ¼ d t 10T 60=5:5 0
0
¼ Pasteurisation units ðPU60 Þ,
ð1Þ
where T is the temperature (C) at the measurement point at any time during the heat process, dt is the duration of time (min) at a particular temperature and tp is the time (min) at the end of the heating process. From this data a second RF cooking protocol (RF2) was developed (by extending the holding time) which produced a comparable PU60 to the conventionally cooked samples. This in total four cooking protocols were developed: (1) steam cooked to 73 C for 2 min (33 min at 80 C in an S); (2) WB cooked to 73 C for 2 min (29 min in 80 C in a WB); (3) RF cooked to 73 C for 2 min (7 min 30 s at 450 W RF with 80 C circulating water); and (4) RF cooked to 73 C for 2 min followed by an extended holding time to produce a similar PU60 to that produced in steam cooked samples (7 min 40 s at 450 W RF with 80 C circulating water). 2.6. Instrumental texture measurement An Instron, Universal Testing Machine (Merlin, Model No. 5544, Instron Corporation, Highwycombe, England) was used for all textural measurements in conjunction with the Instron Merlin software package (Version 2019). Textural profile analysis (TPA) was conducted using the methods of Bourne (1978) as described by Colmenero, Barreto, Mota, and Carballo (1995). The Penetration test was performed using the method of Cavestany, Colmenero, Solas, and Carballo (1994). For Kramer shear analysis, slices (2.5 cm ·
5.5 cm · 4 mm) were prepared using a template. These samples were immediately wrapped in cling-film and allowed to equilibrate to room temperature. The WB analysis was performed at a crosshead speed of 200 mm min1 and both load and extension were analysed. For each of the textural analyses conducted, puddings from three separately prepared batches were assessed with two puddings taken per batch and six readings taken from each pudding. 2.7. Expressible fluid test (EFT) Preweighed meat cores (2.5 cm diameter, 2 cm high) were prepared following cooking and cooling of the meat emulsion batter. Cores were immediately wrapped in cling film and allowed to equilibrate to room temperature for 2 h. Cores were then removed from cling film, placed between three sheets of preweighed Whatman #1 filterpaper (9 cm diameter) with two sheets underneath and one above the core. Samples were compressed to 75% of their total height using an Instron (Model No. 5544, Instron High Wycombe, Bucks, UK) operating at a crosshead speed of 50 mm min1. Total expressible fluid (TEF) consisting of expressible moisture and expressible suspended solids were calculated using the methods of Lee, Whiting, and Jenkins (1987) and Cavestany et al. (1994). For each of the textural analyses conducted, puddings from three separately prepared batches were assessed with two puddings taken per batch and six readings taken from each pudding. 2.8. Colour measurement Colour measurements were performed on cores prepared for TPA, prior to TPA measurement. A chroma meter (Model CR-300 Minolta, Minolta (UK) Limited, 1-3 Tanners Drive, Blakelands North, Milton Keynes, MK14 5BU, England) was used to determine Hunter L* (lightness), a* (redness/greenness) and b* (yellowness/blueness) from which the hue angle (H) and the saturation (S) were calculated using the equations give by MacDougall and Rhodes (1972). The chroma meter was calibrated for internal light (D65) before carrying out colour measurements. As readings were taken from TPA and penetration test samples, puddings from three separately prepared batches were assessed with four puddings taken per batch, six cores per pudding and two readings per core. 2.9. Sensory analysis A triangle taste test was conducted to evaluate whether panellists can distinguish between WB cooked samples and samples prepared by RF2. Untrained panellists were recruited to carry out the evaluation. Prior to each session samples were cut into 20 mm thick slices
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
and grilled using a domestic cooker for 500 s. Each panellist was given three samples (two identical and one different) of cooked WP and asked to record the number of the odd sample. A separate batch of puddings was prepared for sensory analysis and testing was continued until a significant result was obtained.
341
Table 1. When ANOVA was conducted on each of the attributes assessed, no significant differences between the TPA, Kramer shear or penetration test values of products cooked by the different methods was observed (P > 0.05). 3.3. Expressible fluid analysis
2.10. Statistical analysis One-way analysis of variance (ANOVA) was used to test the effect of cooking protocol on the quality evaluation assessed using the SAS statistical analysis software (Version 8.2, Statistical Analysis Systems, Cary, NC, USA). For sensory analysis a sequential test as described by Meilgaard, Civille, and Carr (1991) was used to analyse the data.
3. Results 3.1. Temperature profiles and lethality Fig. 2 shows a time–temperature profile for a cased WP cooked using a heated circulating water cell. A time temperature profile for an identical pudding heated using a steam-oven is provided for comparison. A standard cooking time of 7 min 30 s was required to achieve a temperature of 73 C while for steam cooked samples this was only achieved after 33 min. Whilst both RF and steam cooked samples achieved similar end-point temperatures, a further 10 s holding time was required for RF cooked samples receive an identical lethality to the steam cooked samples. 3.2. Instrumental texture A range of textural parameters for WP cooked using RF or conventional cooking protocols are presented in
Results of Expressible fluid analysis of WP cooked using RF or conventional cooking protocols are presented in Table 2. No significant differences in yield expressible fluid (%) and expressible moisture (%) were recorded between products cooked by the various cooking methods (P > 0.05). 3.4. Instrumental colour Results of instrumental colour analysis of WP cooked using RF or conventional cooking protocols are presented in Table 3. No significant differences in colour (i.e. hue angle, saturation, L, a* and b* values) were recorded between products cooked by the various cooking methods (P > 0.05). 3.5. Sensory analysis Results for the sequential test are plotted in Fig. 3. Equations provided by Meilgaard et al. (1991) were used to plot lines on this graph with number of panellists on the x-axis and number of correct tests on the y-axis. These lines divide the plotted area into three regions: the acceptance region (i.e. accept that samples are similar), the rejection region (reject that samples are similar), and the continuing testing region. Testing was continued until the plotted line crossed one of the lines bordering the continue testing region. In this work it was found necessary to use 57 panellists before it could be concluded that no perceivable difference exists between the samples (P > 0.05).
90
4. Discussion
80
Temperature (oC)
70 60 50
RF
40
S
30 20 10 0 0
5
10
15 20 Time (min)
25
30
35
Fig. 2. Typical time–temperature profiles of radio frequency (RF) and steam (S) cooked white pudding.
RF heating offers the advantage of a reduction in processing times over conventional methods. However, any advantage conferred on meat processors by this reduction would be lost if application of the technology resulted in a loss of product quality. In this light the lack of any significant differences (P P 0.05) in the textural and colour parameters between RF1, RF2 and conventionally processed WPÕs (Table 1) suggests this technology looks promising for this particular product. In addition, triangle tests of WP cooked by RF2 and steam cooking methods indicated that panellists could not differentiate between the conventionally and RF cooked samples. In contrast Van Roon et al. (1994) reported
342
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
Table 1 Texture of white pudding cooked by water bath (WB), steam (S) and radio frequency (RF) methods Instrumental texture attribute
RF1a
RF2b
Mean Kramer shear Load (N) Extension (mm) Texture profile analysis Chewiness (N mm) Cohesiveness Cohesion force (N mm2) Springiness (mm) Hardness1 (N) Hardness 2 (N)
b c d
Mean
SD
WBd
Mean
SD
Mean
P-value SD
86.4 7.61
11.05 3.238
84.5 7.55
16.492 3.326
86.2 8.054
13.928 3.905
89.4 8.03
12.374 3.887
0.972 0.996
194.4 0.515 77.8 7.75 48.9 37.0
25.65 0.041 1.693 0.161 3.174 3.124
177.9 0.503 76.9 7.69 45.5 34.7
41.94 0.054 2.378 0.239 7.587 8.223
223.2 0.516 79.6 7.94 54.2 41.3
53.37 0.082 1.655 0.160 8.537 9.207
203.3 0.482 79.6 7.94 53.5 38.0
31.52 0.063 0.755 0.075 4.366 6.203
0.577 0.835 0.726 0.723 0.467 0.622
2.89 5.96
0.609 0.389
2.75 6.06
0.427 0.383
2.94 5.86
0.380 0.291
2.79 5.47
0.394 0.684
0.895 0.198
Penetration test Penetration test load (N) Penetration elastic behaviour (mm) a
SD
Sc
RF1: 450 W for 7 min 30 s with 80 C circulating water followed by 2 min holding time in 80 C water. RF2: as RF1 but holding period extended by 10 s. S: cooked for 33 min at 80 C in a steam oven. WB: cooked for 29 min in water at 80 C water.
Table 2 Yield and expressible fluid of white pudding cooked by water bath (WB), steam (S) and radio frequency (RF) methods Instrumental quality parameter
Yield (%) Total expressible fluid (%) Expressible moisture (%) a b c d
RF1a
RF2b
Sc
WBd
P-value
Mean
SD
Mean
SD
Mean
SD
Mean
SD
0.999 0.019 0.833
0.004 0.004 0.03
1.000 0.018 0.837
0.001 0.006 0.028
1.000 0.019 0.832
0.001 0.005 0.028
1.000 0.018 0.807
0.001 0.004 0.044
0.342 0.980 0.698
RF1: 450 W for 7 min 30 s with 80 C circulating water followed by 2 min holding time in 80 C water. RF2: as RF1 but holding period extended by 10 s. S: cooked for 33 min at 80 C in a steam oven. WB: cooked for 29 min in water at 80 C water.
Table 3 Colour attributes of white pudding cooked by water bath (WB), steam (S) and radio frequency (RF) methods Instrumental colour attribute
L* Hue angle () Saturation a* b* a b c d
RF1a
RF2b
Sc
WBd
P-value
Mean
SD
Mean
SD
Mean
SD
Mean
SD
61.468 47.465 9.366 6.322 6.898
3.147 2.845 0.931 0.667 0.792
61.485 47.412 9.301 6.283 6.848
3.175 2.53 0.941 0.624 0.813
61.671 46.496 9.393 6.466 6.804
3.482 2.43 1.003 0.794 0.727
61.856 47.584 9.422 6.340 6.955
3.634 3.2 1.075 0.739 0.933
0.997 0.917 0.977 0.977 0.992
RF1: 450 W for 7 min 30 s with 80 C circulating water followed by 2 min holding time in 80 C water. RF2: as RF1 but holding period extended by 10 s. S: cooked for 33 min at 80 C in a steam oven. WB: cooked for 29 min in water at 80 C water.
that RF cooked meat doughs were much firmer than WB cooked samples. In addition, Laycock, Piyasena, and Mittal (2003) reported that RF cooked comminuted beef samples had significantly lower hardness 1 and hardness 2 values indicating that the RF product was more ÔsoggyÕ than the conventionally cooked product.
The products described by Laycock et al. (2003) were comminuted beef with no other added ingredients. Therefore they did not contain salt or other added binders, which would have helped them to retain water. Furthermore, they would not have been representative of typical commercial meat doughs unlike the products in
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
343
Fig. 3. Sensory analysis test plot of results of radio frequency cooked vs. steam cooked white pudding.
this work. These authors also reported differences in springiness and gumminess between RF and WB treated samples. In this work RF cooking resulted in a reduction of cooking times of up to a third when compared to steam cooking. Laycock et al. (2003) reported that a reduction in cooking time of up to 10 fold could be achieved for RF cooked comminuted beef as compared to WB cooked samples. The workers (Laycock et al., 2003) reporting the greater reduction in cooking time used a substantially higher RF input power (696 W versus 450 W in the present case) and thus a relatively greater reduction in cooking time would be expected. The RF input power in the present study was limited by the maximum power output of the RF generator (600 W). Another contrast between previously reported work and the present study was that in this work, similar to the practice used in commercial meat processing operations, the WPÕs (for both RF and conventional cooking) were sealed in plastic casings which ensured that the product maintained a uniform shape during processing. In addition, however, casings help to prevent post-process contamination and eliminate evaporation during cooling and loss of juices, thereby encouraging gel formation and moisture retention in the product. In contrast, packaging of the product used by Laycock et al. (2003) differed between conventional and RF cooking. Products for conventional cooking were packed in a plastic bag during cooking whereas the RF cooked samples were placed without casing, in a polytetraflouroethylene (PTFE) cylinder between the RF electrodes. A principal objective of the present work was the development of a RF cooking protocol that could be integrated with existing commercial production methods but which would allow cooking to take place in a continuous fashion rather than the normal batch type cooking methods which are predominantly used for these products. The authors of the present study suggest that this could be achieved by placing the product to be heated on a conveyor belt and passing it between RF electrodes
immersed in a heated WB. Once the product has been heated to a suitable end-point temperature by application of RF heating, a subsequent holding phase could be achieved by passage of the product through a heated WB without electrodes.
5. Conclusion RF cooking of products in air was found to be unfeasible and therefore it was necessary to surround products with heated water during RF cooking. A cell made from high-density polyethylene, which held the product and allowed water circulation facilitated successful heating of the product. This study has demonstrated that 450 W of RF power could reduce cooking times to 7 min 40 s as compared to 33 min for conventional cooking. Instrumental quality indices of the resultant product revealed no significant differences (P P 0.05) with regard to texture and colour. Subsequent sensory work confirmed this observation. These results suggested that RF heating could be very suitable for industrial heating of encased WP. However, similar studies would be required on a large range of meat products before this conclusion could be reached. Future work should also compare the effect of higher RF output powers and the resultant shorter cooking times on product quality.
References Anon. (1985). Radiofrequency and the production of food. Baking Today, September, 46–47. Anon. (1995). Cooking with radio frequency. Meat International, 5(5), 10–11. Bengtsson, N. E. (1963). Electronic defroisting of meat and fish at 35 and 2450 mcs-a laboratory comparision. Food Technology, (October), 97–100. Bengtsson, N. E., & Green, W. (1970). Radio frequency pasteurisation of cured hams. Journal of Food Science, 35, 681–687.
344
N.P. Brunton et al. / Food Research International 38 (2005) 337–344
Bourne, M. C. (1978). Texture profile analysis. Food Technology, 7, 62–72. Cavestany, M., Colmenero, F. J., Solas, M. T., & Carballo, J. (1994). Incorporation of sardine surimi in bologna sausage containing different fat levels. Meat Science, 38, 27–37. Colmenero, F. J., Barreto, G., Mota, N., & Carballo, J. (1995). Influence of protein and fat content and cooking temperature on texture and sensory evaluation of bologna sausage. LebensmittelWissenschaft Und-Technologie, 28, 481–487. Houben, J., Schoenmakers, L., van Putten, E., van Roon, P., & Krol, B. (1991). Radio-frequency pasteurisation of sausage emulsions as a continuous process. Journal Microwave Power and Electronic Energy, 26(4), 202–205. Institute of Food Technologists (2001). Overarching principles: kinetics and pathogens of concern for all technologies. In D. E. Weber, F. R. Katz, & C. L. Mattson (Eds.), Journal of Food Science Special Supplement: Kinetics of microbial inactivation for alternative food processing technologies (pp. 16–32). Chicago, IL: Institute of Food Technologists.
Lee, C. M., Whiting, R. C., & Jenkins, R. K. (1987). Texture and sensory evaluations of frankfurters made with different formulations and processes. Journal of Food Science, 52, 898–900. Laycock, L., Piyasena, P., & Mittal, G. S. (2003). Radio frequency cooking of ground, comminuted and muscle meat products. Meat Science, 65, 959–965. MacDougall, D. B., & Rhodes, D. N. (1972). Characteristics of the appearance of meat III. Studies on the colour of meat from young bulls. Journal of the Science of Food and Agriculture, 23, 637–647. Mc Cormick, R. (1988). Dielectric heat seeks low moisture applications. Prepared Foods, 162(9), 139–140. Meilgaard, M., Civille, G. V., & Carr, B. T. (1991). Sensory evaluation techniques (2nd ed.). London: CRC Press. Stumbo, C. R. (1973). Thermobacteriology in food processing. London: Academic Press. Van Roon, P. S., Houben, J. H., Koolmees, P. A., & Van Vilet, T. (1994). Mechanical and microstructural characteristics of meat doughs, either heated by a continuous process in a radio-frequency field or conventionally in a waterbath. Meat Science, 38(1), 103–116.