Subscribe Now Subscribe Today
Research Article
 

Reduction of Drying and Ripening Times During the Italian Type Salami Production



Gustavo Graciano Fonseca, Paulo Sergio da Silva Porto and Luiz Antonio de Almeida Pinto
 
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail
ABSTRACT

Salami is a product which preparation time varies notably in function of some parameters, especially its size. This study carried out in a company to verify in industrial scale the possibility of the reduction of the days of process. Drying curves were done for salami with 60 mm diameter and 240 and 260 mm lengths, corresponding to masses of 776 and 852 g, respectively. In the current conditions, the process time was of 38 days. It was concluded that the maintenance of the 7 days of pre-ripening, a reduction in the period of the first ripening from 13 to 7 days and in the second ripening from 18 to 17, totaling 31 days, reached the desired humidity, occurring no alterations in the final product characteristics. Through non-linear regression analysis it was verified that the obtained parameters of the drying curves, had presented excellent adjustment of correlation (R2 > 99%).

Services
Related Articles in ASCI
Search in Google Scholar
View Citation
Report Citation

 
  How to cite this article:

Gustavo Graciano Fonseca, Paulo Sergio da Silva Porto and Luiz Antonio de Almeida Pinto, 2006. Reduction of Drying and Ripening Times During the Italian Type Salami Production. Trends in Applied Sciences Research, 1: 504-510.

DOI: 10.17311/tasr.2006.504.510

URL: https://scialert.net/abstract/?doi=tasr.2006.504.510

Introduction

Salami is a typical Italian seasoned sausage obtained from different types of pork meat and chopped in several different sizes according to the variety of the salami (Procida et al., 1999). Numerous varieties have been developed based on local traditional recipes leading to products with different compositions, degrees of comminution of the ingredients, casing diameters, seasonings, application of smoking, use of molds and ripening time (Fernández et al., 2000).

The production process depends mostly on the skill and experience of the manufacturer. In the past, dry sausages were manufactured using traditional craft methods and despite increasing industrialization, the continuous processes were developed on an empirical basis. However, both commercial competitiveness and increasing demand in the market for higher quality products have led the need of improving the conditions for ripening (Bello and Sanchez-Fuertes, 1997). Intensive research into ripening process was due to traditional empirical methods of manufacture, not being able to meet the requirements of large-scale low-cost industrial production. In the industrial production, the number of problems rises when is tried to obtain similar products to the traditional ones, with highly standardized forms in short ripening times (Fernández et al., 2000; Procida et al., 1999).

The manufacture of dry fermented sausages is considered to occur in three phases: formulation, fermentation and ripening/drying. The time and energy requirements for the ripening process contribute largely to the total costs of fermented sausage manufacture due to its lengthy duration. Therefore, acceleration of the ripening would result in a reduction of the storage time and would increase the profit margin and the competitiveness of the end product (Fernández et al., 2000). Different strategies have been developed to reduce ripening/drying time. The first assays for accelerating the production were addressed to shorten the drying process, through the removing of as much water as possible prior to fermentation by freeze-drying of the meat, reduction of the water binding capacity of the better by using PSE meat and drying under vacuum (Pardi et al., 1994). These approaches certainly reduce the drying time, but flavor development can be affected, resulting in a less intensive or unbalanced flavor due the shorter processing time.

Salami quality depends on the quality of the raw materials and the technology of production. Thus, a variety of processing practices have been developed using different raw materials, ageing times and manufacturing techniques (Toldrá et al., 1997; Virgili et al., 1997), chemical and enzymatic mechanisms during maturing (Careri et al. 1993; Buscalihon et al. 1994) and start cultures (Geisen et al., 1992; Nuñez et al., 1996; Toledo et al., 1997). Moreover, most of the sensory properties of dry fermented sausages can be attributed to proteins and lipids, or to compounds produced by the further degradation of these products (amino acids and free fatty acids) during the ripening process. For this reason several authors have attempted to accelerate the ripening process by adding proteases and lipases (Blom et al., 1996; Díaz et al., 1996, 1997; Fernández et al., 1995; Melendo et al., 1996).

However, the easiest and safest way to reduce ripening times is through the optimization of the production conditions. Moreover, by an adequate control of the process, the probability of damaged product that can compromise its acceptability is reduced. Thus, the present study was carried out with the objective to verify in industrial scale the possibility of the reduction the process time of two different varieties of Italian type salami produced in a Brazilian plant by the modification of the permanence time of salami in determined ripening operational conditions.

Materials and Methods

The present study was conducted in one of the largest food companies and exporters in Latin America, located in the State of Rio Grande do Sul, Brazil, during the year of 2005.

Processing Stages
Processing includes diverse operations: preparation and cooling of raw materials, mixing, inlaying and drying/ripening (divided in three distinct phases), embedding and packing.

Preparation, Mixture and Embedding of the Mass
After a minimum 24 h cooling time for swine meat and 48 h unfreezing and cooling time for bovine meat, the raw materials were milled. The cooling of the meat was made among 2-7°C. The meat was cut in wide straps to increase the surface in contact with air to favor a certain loss of humidity. The milling of meat was made in 5-8 mm for swine and 3-5 mm for bovine one and in 5-10 mm knives for lard milling that was not cooled to make the work easier. Milled meat portions were weighted and mixed with condiments and additives. Each batch of Italian type salami was necessarily constituted by bovine meat, swine meat and lard. Salt, sugar, GDL (glucone-delta-lactone), ripening salts, dry milk and casein were used as condiments.

From the mixer, the batch was conducted to the inlaying equipment. The salami desired size was regulated and artificial gut coil added to start the inlay. The gut received a previous treatment that consists of diving it into a tank with warm water (40°C). To each gut coil 25 g of Delvocid (Degussa BioActives) was added. The salami was placed in wooden laths and then taken for the drying/ripening chambers.

Drying and Ripening
During the drying/ripening time, salami crosses through three distinct stages: The first one, pre-ripening, during 7 days at 16-18°C temperature range and 86-92% relative humidity; the subsequent one, the first stage of ripening, during 13 days at 10-12°C temperature range and 82-88% relative humidity and the last one, the second stage ripening during 18 days at 12-14°C temperature range and 76-80% relative humidity.

Table 1: Operational conditions in the pre-ripening chambers
Image for - Reduction of Drying and Ripening Times During the Italian Type Salami Production
a,b: Steps with same Relative Humidity (RH) and temperature conditions, but different air speeds

Table 2: Operational conditions in the ripening chambers
Image for - Reduction of Drying and Ripening Times During the Italian Type Salami Production
RH: Relative Humidity

The air speed varied from 0.1 to 0.8 m sec-1. The stages were carried out in cooled chambers, called pre-ripening and ripening chambers. Table 1 and 2 show the operation conditions for pre-ripening and ripening (first and second stages) chambers.

After drying/ripening stages, salami was washed with water, through an aspersion with coils, to remove the developed mildew. After 24 h of drying, salami was covered with vegetal oil and anti-mildew (potassium sorbet solution 2.5%).

Analytical Procedures
Drying curves were obtained utilizing triplicate salami samples. Salami had 240 and 260 mm length and 60 mm of diameter corresponding to 776 and 852 g, respectively. The preparation, mixing and inlaying of the mass had been carried out as described in previous section. The drying/ripening was carried out in three distinct stages, according to Table 1 and 2.

The samples were weighed every 8 hours in pre-ripening, every 12 h in the first stage ripening and every 24 h in the second stage ripening. The produced salami possessed 56.5% initial wet basis moisture content and the desired final moisture content is 35%. It was determined according to AOAC (1990) method. In the current operational conditions (38 days time process), the salami average final moisture content was 33.7%. Drying curves were determined from experimental data and fitted applying statistical analysis with non-linear regression through the least squares method.

Results and Discussion

Experimental data were fitted using a polynomial model. It was evidenced that the third order model reached the best correlation (R2). Figure 1 and 2 show the drying curves fitted to a third order model for the samples of 240 and 260 mm length, respectively, whose experimental assays were obtained in the current industrial conditions. According to Fig. 1, the experimental data for samples with 240 mm length follow the model until approximately the thirteenth day; after that they diverge. In Fig. 2 the same behavior can be observed for samples with 260 mm length. Physically, the explanation for this lump in the drying curves is that the process conditions are not more able to promote the dehydration of the product due the existence of a dynamic equilibrium among humidity of the product and environment (Treybal, 1981). The sample reaches the equilibrium with the 82 to 88% relative humidity and 10 to 12°C temperature.

Image for - Reduction of Drying and Ripening Times During the Italian Type Salami Production
Fig. 1: Drying curve fitted to a third order model for the samples of 240 mm length/current conditions

Image for - Reduction of Drying and Ripening Times During the Italian Type Salami Production
Fig. 2: Drying curve applied to a third order model for the samples with 260 mm length/current conditions

Image for - Reduction of Drying and Ripening Times During the Italian Type Salami Production
Fig. 3: Drying curve fitted to a third order model for the samples with 240 mm length/proposed conditions

Image for - Reduction of Drying and Ripening Times During the Italian Type Salami Production
Fig. 4: Drying curve applied to a third order model for the samples with 260 mm length/proposed conditions

Table 3: Models and correlations for the two Italian type salami standards and conditions (current and proposed)
Image for - Reduction of Drying and Ripening Times During the Italian Type Salami Production

Therefore, a reduction of the first stage ripening time to 7 days was proposed. According to the results, 6 days in the first stage ripening would be enough, but a safety margin was selected. Analogously, Fig. 3 and 4 indicate the drying results obtained for salami samples with 240 and 260 mm length, respectively, whose experimental assays were obtained according to the proposed conditions. As it could be observed, both situations presented a constancy in the humidity reduction, without occurring points out of the model, what is physically justified by the existence of drying during all the process, without periods of stagnation. This linear drop was also observed by Cavenaghi (1999) and Garcia et al. (2000) during salami manufacturing.

Table 3 shows the models and correlations obtained for the two Italian type salami standards (240 mm/776 g and 260 mm/852 g) and the two different conditions (current and proposed). Comparing the two conditions, the experimental data from the proposed conditions presented a better correlation coefficient (R2) when fitted to the polynomial curve, what suggests its higher affinity to the model.

The ripening of inlayed salami is carried out through very different criteria depending on the typology of the products (Pardi et al., 1994). Salami, if correctly ripened, is a steady product, but notable modifications can occur on flavor, mainly related to excessive losses of water and transformations of the adipose tissue, that can be easily oxidized at the surface (Môheler, 1980). Thus, special care must be taken to maintain the desirable characteristics of the final product. The use of controlled acclimatization allowed the attainment of products with high standardization profile, what was observed by the behavior of the triplicate samples, for each studied situation, supplying sufficiently adjusted models. The maintenance of the temperature of pre-ripening in the range of 16 and 18°C and relative humidity between 86 and 92% revealed convenient for fast salami dehydration.

It is important to keep the relative humidity high in the first days (90%) to prevent the extreme drying of the surface and controlled during the following stages, until reaching the desirable color, acidification and final moisture content of the product (Pardi et al., 1994). The proposed process alterations on the first and second stages of ripening did not provoke neither physical nor sensorial alterations in the final product. Two types of problems of physical order referring to an inadequate technological process of salami drying were avoided: incrustation, characterized by an internal barrier (crust) that forms due the fast moisture loss which hinders the adequate drying of the internal section of the salami, or wrinkling, which is the deformation of the salami caused by moisture loss.

According to Brazilian Legislation, the maximum humidity content for salami is 35% (Brazil, 2000). However, Cavenaghi and Oliveira (1999) analyzed six commercial marks of salami from Brazil and determined humidity contents varying from 30.9 to 37.1%. From the economic point of view, it is important to keep this parameter very well controlled to avoid neither loses in the product mass (due moisture loss more than necessary) nor injury the consumer (that buy a product with more moisture instead of meat) and consequently suffer legal penalties.

A modification of the currently used salami ripening process was proposed and analyzed in this work: the maintenance of the 7 days pre-ripening, with a reduction in the first stage ripening from 13 to 7 days and in the second stage from 18 to 17, totaling 31 days. These alterations allowed moisture content reduction in the analyzed samples with no collateral effects on the characteristics of the final product. The current 38 days protocol produces salami with final average moisture content of 33.7%, which means an extra loss of 1.3% of humidity.

It is believed that through the observation of conditions already used the industries allied to simple transport phenomena tools are the still more efficient way to reduce the ripening time of inlayed salami. This way, damage risks in the final quality of the product are eliminated by a rigid control in the process conditions.

©  2022 Science Alert. All Rights Reserved