Viscoelastic Properties of Wiener Sausages During Cooking

Abstract

Viscoelastic properties of Wieners during cooking were measured by a rheometer using a stress creep procedure (stress of 5 Pas and time of 60 s) for the retardation and the recovery stage. Experiments were performed at 6 temperatures (40–65°C) and 4 times (5–14 minutes). The viscoelastic behavior of Wieners was well described (R² > 0.98) by a Burgers model. Main parameters, the instantaneous compliance J_o; the retardation compliance J₁, the retardation viscosity μ₁ and the Newtonian viscosity μ_o, varied significantly with cooking temperature and time. Four second order regression models were developed for these parameters. Predicted and experimental values were in good agreement (R² > 0.84).

Keywords:

• Viscoelastic properties
• Sausage
• Cooking process
• Compliance
• Viscosity
• Stress creep procedure

INTRODUCTION

Textural and viscoelastic properties are important factors in the overall quality of meat sausages and play a vital role in their consumer acceptance.[1] Wiener sausages—made of homogenized meat particles, starch, and other ingredients—are one of traditional meat products commonly sold on the market today. Cooking and smoking processes are necessary to ensure the microbial safety, to obtain desired quality characteristics (e.g., color, taste, and texture), and to extend the shelf life of these products. During these operations, raw sausage emulsions will gelatinize exhibiting typical viscoelastic behavior that will be significantly affected by many factors including processing conditions (e.g., cooking time and temperature) and composition (i.e., mainly the meat protein type, the fat to protein ratio, the salt concentration, and the moisture content).

Studies on gelation and viscoelastic properties of sausage emulsions have been widely reported in the literature. The effect of cooking temperature on textural and viscoelastic characteristics of frankfurters was determined by Singh.[2] They found that the hardness and the compression energy of the first bite increased with cooking temperature. Polynomes were established for cohesiveness, elasticity, gumminess, and chewiness as a function of cooking temperature, with minimum values in the range of 70–75°C. From their study on the mechanisms of gel formation by proteins of the muscle tissue, Ziegler and Acton[3] reported that the heat induced gelation of muscle proteins was largely responsible for the physical and chemical stabilization of fat and water in comminuted red meat products. Textural and viscoelastic properties of meat emulsions were investigated by Siripurapu, Mittal, and Blaisdell,[1] under different fat-protein ratios and cooked at various temperatures and for various times. Results showed that texture properties of sausages changed with the increase of cooking and holding time and with the decrease of the meat fat-protein ratio. A three-element model, comprising a Maxwell (spring) and a Kelvin unit in series, was found to be suitable to describe the viscoelastic behavior under a stress creep test. Studies on the gelation kinetic of meat emulsions containing various fillers and during cooking revealed that transition temperatures existed during cooking. Muscle proteins insolubilization and solubilization of collagen occurred.4–6 The effect of starch properties on viscoelastic characteristics of meat products was examined by Li and Yeh.[7] These authors found that the addition of starch resulted in a decrease in cooking loss and in an increase in both the storage modulus (G') and the loss modulus (G”). Adding starch also reduced the leaching out from meat protein. Viscoelastic properties of Bologna sausages were studied using oscillatory dynamic tests by Bruno and Moresi.[8] Results showed that two constant parameters such as the relaxation function and the stiffness could be used to characterize the protein network of different Bologna sausages.

Most literature studies focused on the development of measurement methods for assessing the viscoelastic behavior of meat emulsions,8–10 or on the compositional effect on viscoelastic properties.11–15 Few studies[1] dealt with the effect of cooking conditions, specifically for assessing the kinetic changes of viscoelastic properties. Kinetic information of viscoelastic properties during the cooking process is necessary for the development of prediction models for quality.[16] The objective of this study was to select a rheological model suitable to describe viscoelastic properties of Wiener sausages using a stress creep testing, to investigate the effect of cooking conditions, and to elaborate related kinetics models as a function of cooking time and temperature.

METHOD AND MATERIALS

Source of Raw Materials

Uncooked raw emulsions of Wiener sausages were obtained from a local company and stored in a refrigerator kept at a temperature of 0–2°C before experimentation. The composition of the emulsion consisted of water (60.51%), fat (20.90%), salt (2.43%), proteins (12.4%), carbohydrates (2%).

Cooking Processes

Each cooking process was directly carried out in the rheometer AR-1000N equipped with a Peltier plate as a pre-step prior to the measurement of viscoelastic properties. The sausage emulsion sample was heated on the bottom plate with a cover to avoid the moisture loss during cooking. The desired heating temperature and holding time (cooking condition) were controlled by a microcomputer. This procedure was advantageous for two reasons. It was convenient to apply the heat directly using the rheometer for each cooking condition prior to perform the actual viscoelastic property measurement. A very uniform cooking temperature was obtained for the whole sample since its thickness (1000 μm) was determined by the distance between the cone-plate and the bottom plate. Six levels of cooking temperatures (from 40°C to 65°C) and 4 levels of cooking time (from 5 to 14 minutes) were used. A full factorial experimental design was applied with three repetitions.

Viscoelastic Property Measurement

Dynamic viscoelastic studies were performed at ambient temperature using a control-stress rheometer (TA Instrument AR–1000N). General conditions for all tests included the use of a cone-plate (cone angle 2°, diameter 2 cm) and a gap of 1000 μm. A dynamic test method, i.e., a creep procedure with a stress of 5 Pas and a time of 60 s for both the retardation and the recovery stage was used for all samples.

Choice of the Mechanical Analogue

Some preliminary trials on the model selection were performed. From the results of correlation coefficients applying regression models it was found that the four-element Burgers model (Fig. 1) was better than the three-element model used by Siripurapu et al.[1] Thus, the four-element Burgers model was selected to describe the viscoelastic properties of Wiener sausages. Mathematically, it is given by:

where J_o is the instantaneous compliance, J₁ is the retardation compliance, μ₁ is the viscosity of the Kelvin component, and μ_o is the Newtonian viscosity of the free dashpot.

Figure 1 Burgers model.

Kinetic Modelling of Viscoelastic Properties

Kinetic models of the above four viscoelastic parameters (J_o, J₁, μ_o, μ₁) as a function of cooking time and temperature were developed using a second order response surface method, which was given as below:

where y is one of the viscoelastic properties, b is a constant value, and x represents one of input variables either time or temperature. Subscripts 1 and 2 represent cooking temperature and time, respectively.

RESULTS AND DISCUSSION

Viscoelastic Behaviour of Wiener Sausages

Fig. 2 shows the viscoelastic behavior of Wiener sausage emulsions under two different cooking conditions. The whole test consisted of two periods: the retardation (t₁) and the recovery (t₂). It can be found that there was a similar trend for the relationship between the compliance and the time for different samples with different processing conditions. During the retardation in which the stress was kept constant, there was an instantaneous change (J_o) in the compliance. It was followed by an increase within which the compliance increased with the retardation time. In addition, the increase stage can be divided into two periods i.e. the t₁₁ and t₁₂. For the period t₁₁, the compliance increased with time but the increasing rate decreased with time. The value of t₁₁ was dependent on the viscosity μ₁. For the period t₁₂, the compliance increased with time linearly and the increasing rate was determined by the viscosity μ₀. During the recovery in which the stress was removed, there was also an instantaneous recovery for the compliance. The recovery rate was then decreasing with time until it reached a plateau. A permanent deformation remained related to the viscoelastic value μ_o. However, by comparison of the compliance curves from different processing conditions, it was obvious that different relationships between the time and the compliance would be obtained. Viscoelastic properties of Wiener sausages were affected by the cooking temperature and time. For example, under the same cooking time of 8 minutes, the compliance with a higher cooking temperature showed a lower value than that with a lower cooking temperature. On other hand, under the same cooking temperature of 45°C, the longer cooking time resulted in a lower compliance value. The Burgers model was applied on experimental data to extract the four parameters of Eq. (1) namely, J_o, J₁, μ_o, μ₁. Statistical results showed that the model predicted very well all experimental curves (R² > 0.98).

Figure 2 Viscoelastic behavior performed at ambient temperature using a stress creep procedure.

Effect of Cooking Time and Temperature on Viscoelastic Properties

Values of viscoelastic properties for the Wiener sausage samples, processed with the same cooking time of 5 minutes but varied cooking temperatures, are shown in Fig. 3a. It was found that the compliance value J₁ decreased with the cooking temperature, but the decreasing rate was dependent on the temperature range. For example, when the temperature increased from 40 to 55°C, the value of J₁ was changed from 1.2E-3 to 2.3E-4 m²/N, while when the temperature increased from 55 to 65°C, it was decreased to 1.5E-4 m²/N. The compliance J_o increased with the cooking temperature in the lower temperature range but decreased at the higher temperature range. The increase of the cooking temperature resulted in the decrease of the compliance. Consequently, the elastic module increased with the cooking temperature. Curves for viscosity coefficients μ_o and μ₁ showed that the effect of cooking temperatures was different. For μ_o, it decreased with the cooking temperature during the range of 40 to 50°C but increased within the range of 50–65°C. For μ₁, it increased with the cooking temperature, although it was not obvious on the graph because it was much smaller than the value of μ_o and both of them used the same y-axis.

Figure 3 Individual effect of cooking conditions on viscoelastic properties.

Fig. 3b illustrates experimental results of viscoelastic properties for Wiener sausage samples, processed with the same cooking temperature of 50°C but with different cooking times. It showed that the cooking time affected the viscoelastic properties considerably, except for the μ₁ which is the viscosity of the Maxwell model. However, the effect of cooking times on different viscoelastic properties was also different. Both compliances (J₁ and J_o) decreased with the cooking time, especially J₁. Viscosity parameters μ₁ and μ_o increased with the cooking time, but it was not very apparent graphically. Both cooking temperature and time affected the viscoelastic property values differently. It can also be deduced that relationships between viscoelastic properties and cooking time and temperature were non-linear.

Kinetic Modelling of Viscoelastic Property Parameters

Previous results can be used to reveal individual effects of both cooking temperature and time on viscoelastic properties, but they cannot explain the combined effect of these two parameters on viscoelastic properties. From a practical stand-point, it is more important to develop kinetic models for viscoelastic properties, which can describe the combined effect of cooking time and temperature. Generally, a first order model is used to model the effect of cooking time, while an Arrhenius model is used for the effect of cooking temperature. However, it has been demonstrated earlier that a first-order kinetic model would not appropriate to model the effect of cooking time and temperature on viscoelastic properties. Therefore, a second order response surface model was used for the regression model to fit experimental data (listed in Table 1) between viscoelastic properties and cooking conditions (Eq. 2).

Table 1 Calculated results of viscoelastic parameters from experimental data under various processing conditions.

Temperature (°C)	Time (min)	J0	J1	μ0	μ1
40	5	2.50E − 04	1.25E − 03	4.09E + 05	4.24E + 04
45	5	3.10E − 04	6.50E − 04	2.96E + 05	4.06E + 04
50	5	4.15E − 04	6.40E − 04	1.47E + 05	4.19E + 04
55	5	2.90E − 04	2.90E − 04	3.54E + 05	5.08E + 04
60	5	2.50E − 04	1.60E − 04	4.80E + 05	5.63E + 04
65	5	1.85E − 04	1.50E − 05	6.35E + 05	6.32E + 04
40	8	3.20E − 04	8.00E − 04	2.90E + 05	3.20E + 04
45	8	4.30E − 04	4.50E − 04	1.00E + 05	3.50E + 04
50	8	3.35E − 04	3.00E − 04	2.57E + 05	4.37E + 04
55	8	2.80E − 04	2.90E − 04	4.17E + 05	5.45E + 04
60	8	2.00E − 04	1.20E − 04	5.59E + 05	6.47E + 04
65	8	1.30E − 05	1.10E − 05	7.63E + 05	7.56E + 04
40	11	3.99E − 04	4.70E − 04	1.64E + 05	2.52E + 04
45	11	3.80E − 04	3.30E − 04	1.41E + 05	3.02E + 04
50	11	3.05E − 04	3.70E − 04	3.34E + 05	4.82E + 04
55	11	2.00E − 04	2.60E − 04	4.70E + 05	6.08E + 04
60	11	1.80E − 04	1.30E − 04	5.84E + 05	6.62E + 04
65	11	1.12E − 04	5.10E − 05	8.90E + 05	8.80E + 04
40	14	4.20E − 04	5.50E − 04	1.25E + 05	2.00E + 04
45	14	3.20E − 04	4.10E − 04	1.70E + 05	2.44E + 04
50	14	2.35E − 04	2.10E − 04	4.18E + 05	5.24E + 04
55	14	2.47E − 04	2.00E − 04	5.26E + 05	6.31E + 04
60	14	1.20E − 04	5.00E − 05	6.34E + 05	8.09E + 04
65	14	3.10E − 05	8.21E − 05	7.52E + 05	1.23E + 05

Using multiple regressions, four kinetic models as a function of cooking time (t) and temperature (T) were developed for J₁, J_o, μ₁ and μ_o, respectively. Table 2 shows the analysis of variance related to each parameter. Some parameters can be eliminated because of their non-significance. The interaction between T and t was included in all of them, meaning that viscoelastic properties were significantly affected by the time-temperature interaction. In addition, it can be found that for the cooking temperature, both the linear and the second order terms exhibited significant effects on viscoelastic properties except for J₁. For the cooking time, only the linear terms had significant effects on viscoelastic properties except for J₁. For this parameter, the second order term for t needed to be added. Deleting the non-significant terms (p > 0.05), final regression equations for each property were obtained:

The modeling performance of the above four regression models is shown in Figure 4. Both visual and statistical results indicated that predicted values from the models can match very well experimental results at a significance level of modeling fitness of p < 0.05. The selected second order regression model was adequate for describing the relationships between cooking conditions and viscoelastic properties.

Table 2 Analysis of variance of parameters.

	Probability of significance level
	Jo	J1	μ0	μ1
T	0.029	0.0003	0.005	0.048
T	0.006	0.001	0.002	0.000001
T*t	0.004	0.007	0.002	0.00016
T*T	0.042	—	0.007	0.0421
T*t	—	0.0003	—	—

Figure 4 Modeling performance of second order regression equations.

Prediction Effects of Combined Cooking Time and Temperature

Figs. 5a–d show the combined effect of cooking time and temperature on viscoelastic properties including J₁, J₀, μ₁ and μ₀. As compared to Figure 3 for the individual effect, these 3D graphs can give much more information about both effects of temperature and time on viscoelastic properties. From Figure 5a, it can be seen that under a given range of temperatures, the effect of cooking time on J_o was dependent on the cooking temperature. For example, when T = 60 °C, the J_o decreased with the cooking time, while when T = 40°C, it increased with the cooking time. Similarly, the effect of the cooking temperature was dependent on the cooking time. When t = 4 min, changing cooking temperature did not result into a significant effect on the J_o, but if t = 14 min, the J_o was decreased significantly with the cooking temperature. It confirmed that the effect of the interaction between T and t on J_o was significant. From Figure 5b, it was observed that there was a different effect of processing conditions on J₁ as compared to J_o. First, the effect of temperature was more significant for a short cooking time than that for a long cooking time. For example, when t = 4 min, the J₁ decreased from 0.00115 to 0.0002 m²/N, with the cooking temperature increased from 40 to 60°C, while when 5 = 14 min, the J₁ was only changed from 0.00047 to 0.00016 m²/N. Second, there was a negative effect on J₁ when the cooking temperature was low, but at higher temperature like 60°C, the J₁ decreased with time at the early stage and then increased with time at the later stage. Figure 5c,d showed that there was a similar trend for the effect of cooking time and temperature on the viscosities μ_o and μ₁. At first, μ₁ and μ_o decreased with the cooking time when T was 40°C, while they increased when T was 60°C. Secondly, the effect of cooking temperature on μ_o and μ₁was more significant at t = 14 min than at t = 4 min. It also confirmed that there was a significant interaction between T and t on μ₁ and μ_o.

Figure 5a Combined effect of cooking time and temperature on J_o.

Figure 5b Combined effect of cooking time and temperature on J₁.

Figure 5c Combined effect of cooking time and temperature on μ_o.

Figure 5d Combined effect of cooking time and temperature on μ₁.

CONCLUSIONS

Based on the above study, it can be concluded that the Burgers model can well describe viscoelastic properties of Wiener sausages. Kinetic models of viscoelastic properties were developed as a function of cooking time and temperature using second order regression equations. A significant interaction between cooking time and temperature existed. Results of the modeling performance confirmed there was a good agreement between predicted and experimental data. These models were further used to build 3D graphs in order to predict viscoelastic properties under a range of experimental cooking conditions.

ACKNOWLEDGMENTS

The authors would like to thank Olymel-Flamingo for providing Wiener sausage emulsions to perform these experiments and the Program on Energy Research and Development (PERD) from Natural Resources Canada as well as the Canadian Agricultural Rural and Development (CARD) for providing the funding to perform this research.

Notes

2. Singh, Y. Effect of cooking temperature treatment on the textural and visco-elastic characteristics of frankfurter emulsion. PhD diss., Ohio State University, 1977.

16. Chen, C.R. Application of computer simulation and artificial intelligence technologies for modeling and optimization of food thermal processing. PhD diss., McGill University, 2001.