Instrumental Texture and Sensory Characteristics of Cod Frankfurter Sausages

Abstract

Effects of five ingredients on cod sausages' chemical composition, pH, colour, texture, and sensory attributes were analysed. Changed ingredients were pea protein (0 vs. 2 g/100 g sausage), carrageenan (0 vs. 1 g/100 g), sodium bicarbonate (0, 0.15, and 0.3 g/100 g), sodium chloride (1.2 vs. 2.3 g/100 g) and pork meat (0, 13.3, and 26.7 g/100 g). Sausages became more yellow, springier, and harder with pea protein incorporation. Carrageenan addition made products redder, harder, and springier but less elastic. Sodium chloride reduction produced harder, springier, and more cohesiveness, but it produced less salty and smoky sausages. Textural effects of sodium bicarbonate addition were similar. Pork meat substitution made sausages harder, springier, and more cohesive.

Keywords:

• Cod sausage
• Pea protein
• Carrageenan
• Salt
• Sodium bicarbonate
• Texture
• Sensory analysis
• Colour

INTRODUCTION

Presently, there is a growing interest in developing meat analogues using alternative sources of protein, as a result, health-conscious consumers are promoting the research and development of different meat systems.[1,2] Among these, restructured fish products and new food ingredients, such as vegetable fibres, have been used with the purpose of reaching young consumers. These new products and the possibility of use of wastes may be a solution for the discards of the fish processing industry.[3] Thus, fish can be tested as a meat analogue and several studies have addressed production of attractive fish burgers[4] or fish sausages[5] with promising functional ingredients, such as vegetable or milk proteins and hydrocolloids. Sausages have useful organoleptic and technological characteristics and wide acceptability; thus, they can be specially suited for testing the potential of some fish species. In Portugal, the existence of plenty salted cod waste (Gadus morhua) derived from the production of klipfish (bacalhau) led to the fabrication of a cod Frankfurter sausage containing inner pea fibre, taking a previously developed hake (Merluccius capensis) sausage as model.[6] This cod sausage had some textural problems, particularly, an excessive softness, therefore, requiring a study of the effect of different variables on its quality.

Regarding fish sausages, with total or partial replacement of livestock meat, there are several products developed recently; however, most of them neither totally replace pork meat[7] nor mimic the common Frankfurter sausage.[8,9] Nonetheless, it was found that one sausage formulation with similar quantities of surimi and pork meat could maintain the hardness of commercial sausage without adverse effects on flavour, acceptability and consumer preference.[10] Other authors reported that fish sausage prepared from fish meat and surimi at a ratio 40:60 with 10% pork fat had also acceptable sensory scores.[5]

In the development of new food products, proteins from leguminous seeds are increasingly important because they improve various functional properties, namely, the emulsifying, gelling and water binding capacities.[11] These effects are enhanced by combining these proteins with hydrocolloids.[12] Particularly, it was reported that pea protein combined with a hydrocolloid showed a favourable effect on the texture of a processed food.[13]

Carrageenans, hydrocolloids extracted from algae, have also been used for technological purposes in fish products, for instance, it was found that addition of k-carrageenan up to 2% favoured a better gelation of raw surimi-type paste from horse mackerel (Trachurus murphyi), provided that a moderate thermal treatment was applied (14). Other authors working with giant squid (Dosidicus gigas) found that only addition of ı-carrageenan along with non-muscle proteins enabled a marked improvement in gel formation.[15] On the other hand, regarding low fat beef Frankfurters, increased carrageenan (a mixture of approximately 2/3 k- and 1/3 ı-carrageenan) concentration improved the water holding capacity.[16]

Sodium chloride is a common sausages ingredient, however, a reduction of sodium intake in diets is highly recommended in order to lower blood pressure.[17] Salt is added to meat products in order to enhance their binding and water-holding capacities.[18] Furthermore, in a study with beef sausages was reported that besides saltiness and hardness also flavour intensity scores became higher as salt increased.[19] In spite of the importance of these traits in a commercially acceptable product, it was shown that other factors such as pH can compensate lowered salt concentrations.[18]

Various studies have shown that pH is a fundamental factor conditioning the texture of sausage—namely, a pH increase of the meat batter to 6.0 or higher has been advised in order to overcome product softness.[20] Specifically, the authors have shown that an adjustment of the batter pH from 5.7 to 6.2 increased the hardness and rupture force of ı-carrageenan treated sausages by more than 50%. However, for beef sausages without any incorporated hydrocolloid, a hardness decrease with pH increase (above 6.3 in the manufactured sausages)—probably due to an excessive amount of bound water at higher pH, was reported by other authors,[19] which also concluded that a level of sodium bicarbonate of 0.4% was enough to increase pH from 6.0 to 7.0.

There are significant differences between meat and fish sausages arising from lower total collagen content[21] and higher moisture level in fish.[6] Hence, the replacement of meat muscle in sausages causes a marked decrease in their hardness.[10] For instance, this textural parameter was more than 70% higher in a sausage with equal amounts of pork meat and hake than in a no pork hake sausage.[6] Nevertheless, the hardness of this fish sausage was scored as acceptable by a group of trained panellists.[6]

Therefore, in order to study the possibility of upgrading salted cod wastes, we sought to assess the effect of pea protein, carrageenan, sodium bicarbonate, sodium chloride and pork meat contents on the quality of Frankfurter cod sausages by measuring their chemical composition, pH, colour, instrumental textural, and sensory properties.

MATERIAL AND METHODS

Raw Materials and Additives

Salt-treated, desalted, and quick-frozen cod (Gadus morhua) were bought already cut in pieces (10–15-cm length and about 5-cm thickness) from a local retailer. Each fish batch was kept frozen at –28 ºC and processed within one to two weeks after its arrival at the laboratory. Pork fat and meat were also bought from a local retailer and stored in a refrigerator, no more than two days until processing.

Pea protein (Pisane® M9) was supplied by Cosucra, S.A. (Warcoing, Belgium) and carrageenan (Ceamgel 1830) was obtained from Ceamsa (Porriño, Spain). The protein content of Pisane is 90%, being the remaining 10% mainly carbohydrates and ash. Furthermore, this pea protein isolate presents itself as a cream colour powder whose grains are smaller than 200 μm. Ceamgel 1830 is a mixture of iota and kappa carrageenans extracted from red seaweeds of the class Rhodophyceae, containing additionally locust bean gum, potassium chloride and dextrose. This product presents itself as a pale yellow powder whose grains (98% of them) are smaller than 250 μm.

The other additives were all food grade materials manufactured by different companies: (i) Swelite®, inner pea fibre from Cosucra, S.A. (Warcoing, Belgium); (ii) Potato starch from Emsland-Stärke GmbH (Emlichheim, Germany); (iii) TARIPROT® 1010, emulsifying soy protein from BK Giulini (Ladenburg, Germany); (iv) SOLCON/MAICON 70, soy protein concentrate powder from Solbar Hatzor Ltd. (Ashdod, Israel); (v) Dextropam 100, dextrose from Copam, S.A. (Loures, Portugal); (vi) - VATEL® Salt, sodium chloride from VATEL (Alverca, Portugal); (vii) TARIMIX® Frankfurt, sausage seasoning from BK Giulini (Ladenburg, Germany); (viii) TAROMA® Smoke, smoke aroma from BK Giulini (Ladenburg, Germany); (ix) Sodium bicarbonate from José M. Vaz Pereira, S.A. (Abrunheira, Portugal).

Experimental Design

The effects of different additives and percentages of pork meat substitution on the quality parameters (chemical composition, pH, colour, textural properties, and sensory perception) of Frankfurter cod sausages were evaluated through four experiments prepared respectively from four different cod batches. In a first experiment, the mean quality parameters of three types of Frankfurter cod sausages were determined: with no pea protein and carrageenan (control) (product A); with Pisane, 2 g/100 g sausage and no carrageenan (product B); and with Pisane and carrageenan, 2 and 1 g/100 g (product C), respectively. Positive results led to use these quantities of pea protein and carrageenan in all further trials. In a second experiment, the quality of Frankfurter cod sausages containing 2.3 (control) (product D) and 1.2 g (product E) sodium chloride/100 g was assessed. The favourable effect attained when sodium chloride was reduced to half, led to use the same low salt content in the following experiments. A third experiment involved the measurement of the mean quality parameters of Frankfurter cod sausages containing different levels of sodium bicarbonate, 0 (control) (product F), 0.15 (product G), and 0.30 g/100 g (product H). Afterwards, in a fourth experiment, the same parameters of Frankfurter cod (control) and cod-pork (0, 13.3 and 26.7 g/100 g, i.e., 0, 25 and 50% cod replacement, products I, J, and K) sausages with no sodium bicarbonate were determined. Error assessment was derived from replication of the various analyses performed.

Production of Sausages

Between 4 and 6 kg of frozen cod (salt treated and desalted) were thawed overnight in a refrigerator. Afterwards, skin and bones were manually removed. The resulting cod flesh was minced one single time in a model 84145 meat grinder (Hobart, Troy, OH, USA), equipped with 2-cm grind blades and a metallic screen with 6-mm-diameter circular holes. The appropriate quantities of the various ingredients were weighed in order to produce Frankfurter cod sausage batches of 2 kg each (Table 1). Regarding the preparation of the sausage batters, five sequential steps were always followed:

Table 1 Used sausage recipes

Ingredients	1^st Experiment pisane + carrageenan (g/100 g sausage)			2^nd Experiment salt (g/100 g sausage)		3^rd Experiment sodium bicarbonate (g/100 g sausage)			4^th Experiment replacement by pork (g/100 g sausage)
	Prod. A0 + 0	Prod. B2 + 0	Prod. C2 + 1	Prod. D2.3	Prod. E1.2	Prod. F0.00	Prod. G0.15	Prod. H0.30	Prod. I0	Prod. J25	Prod. K50
Cod mince	54.4	53.4	52.8	52.8	53.4	53.4	53.4	53.3	53.4	40.1	26.7
Pork meat	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	13.3	26.7
Water/ice	25.1	24.6	24.4	24.4	24.7	24.7	24.6	24.6	24.7	24.7	24.7
Pork fat	7.9	7.7	7.6	7.6	7.7	7.7	7.7	7.7	7.7	7.7	7.7
Swelite	3.9	3.8	3.8	3.8	3.8	3.8	3.8	3.8	3.8	3.8	3.8
Potato starch	3.1	3.1	3.0	3.0	3.1	3.1	3.1	3.1	3.1	3.1	3.1
Sodium chloride	2.3	2.3	2.3	2.3	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Pea protein	0.0	2.0	2.0	2.0	2.0	2.0	2.0	2.0	2.0	2.0	2.0
Emulsifying protein	1.3	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Soy protein	1.1	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0
Carrageenan	0.0	0.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0
Sodium bicarbonate	0.0	0.0	0.0	0.0	0.0	0.0	0.2	0.3	0.0	0.0	0.0
Other ingredients	0.9	0.9	0.9	0.9	0.9	0.9	0.9	0.9	0.9	0.9	0.9
Total	100	100	100	100	100	100	100	100	100	100	100

1. Cod mince (in the fourth trial, also pork meat) was mixed with sodium chloride for 1 minute at 1420 rpm in a model UM12 refrigerated vacuum homogeniser (Stephan and Söhne, Hameln, Germany). Throughout all process, mixing was performed always under vacuum and refrigeration (temperature below 7 ºC);

2. Ice (70% of the total amount of ice/water) and dextrose were added and there was additional mixing for 1 minute at the same speed;

3. Emulsifying protein, soy protein concentrate and pea protein (except in the control of the first trial) were also added and further mixed at the same speed for 1 minute;

4. Pork fat was added and mixed for 1 minute once more at 1420 rpm (meanwhile, in order to hydrate Swelite before mixing it with the other ingredients, chilled water was added to this dietary fibre in a ratio near 2:1);

5. Remaining ingredients, potato starch, hydrated fibre, sausage seasoning, smoke aroma and, in some trials, carrageenan and sodium bicarbonate were put in the homogeniser and mixed all for 2 minutes at the same speed.

The batters attained were put inside a model EB-12 hydraulic filler (Mainca Equipamientos Carnicos, S.L., Granollers, Spain) and expelled under pressure through the stuffing horn into cellulose sausage skins mounted over the end of the stuffing horn. Afterwards, these cellulose casings were twisted and tied, thereby, shaping sausages with a diameter of 25 mm and a length of about 20 cm. In the next step, sausages were moved to a model Combi-Master CM6 oven (Rational Grossküchen Technik GmbH, Landsberg am Lech, Germany) equipped with a digital thermometer and subjected to a steam cooking at 75ºC for 15 minutes (cooking). Immediately after, sausages were taken from the oven and cooled with a mixture of water and ice (1:1, v/v). The cellulose skins were removed, sausages separated from one another and vacuum-packed in low-oxygen permeable barrier bags (O₂ transmission, <2.1 cm³/(m².day.bar) at 23ºC, Colamin XX 100e, Obermühle Polymertechnik GmbH, Pössneck, Germany) with a model A300/52 vacuum packager (Multivac Sepp Haggenmüller GmbH & Co. KG, Wolfertschwenden, Germany). Following this operation, sausages were put in the same oven and subjected to a steam cooking at 90ºC for 10 minutes (pasteurisation). In order to guarantee these conditions even in the innermost part of the sausage, the oven's digital thermometer was placed in the centre of a sausage through a hole in the bag. After pasteurisation, packages were immediately cooled in iced water and kept in a refrigerator (2 ± 1ºC) overnight until analysis.

Chemical Composition

Moisture, crude protein, and ash were determined by standard procedures,[22] whereas crude fat was determined by a rapid method of total lipid extraction and purification.[23] All determinations were done in duplicate. The remaining component was considered total carbohydrate.

pH

pH was measured on the cooked sausages using a Sen-Tix 21 pH electrode (Wissenschaftlich-Technische Werkstätten, Weilheim, Germany) on a model pH 539 pH meter (Wissenschaftlich-Technische Werkstätten, Weilheim, Germany).

Colour

Sausages were taken out from packages, cut and put into two Petri dishes, covering the entire bottom. A model MACBETH COLOR-EYE® 3000 colorimeter (Macbeth, New Windsor, NY, USA) was used and, prior to measurements, standardized to a specific colour blank (CIELAB system: L∗, 92.4; a∗, -1.0; b∗, 1.5). The attained values for L∗, a∗ and b∗ of the CIELAB system were always the means of ten measurements on each Petri dish. Furthermore, for a better assessment of colour, the three mentioned coordinates were combined in order to obtain the chroma and whiteness values:

(1)

(2)

Texture Measurements

Folding test (FT)

Sausages were taken out from packages and tempered to about 20°C. The test piece was a 3-mm slice (25-mm diameter) cut from the sausages middle portion. The evaluation was performed in accordance with a 5-point grade system as follows. Grade 5, no crack when folded into quadrants; Grade 4, no cracks when folded in half; Grade 3, crack develops gradually when folded in half; Grade 2, crack develops immediately when folded in half; Grade 1, crumbles when pressed by finger. This test was performed by an experienced technician using surimi samples as reference.

Puncture test

Prior to analysis, sausages were cut into pieces of 25 mm diameter and 25 mm high. Each sample was penetrated to the breaking point with a metal probe equipped with a 5 mm diameter spherical head, using a model Instron 4301 texturometer (Instron Engineering Corp., Canton, MA, USA). The cross speed head was 10 mm/min, load cell was 1000 N, sensitivity of the force measurement was 0.1 N. Breaking force (N), and breaking deformation (mm) were measured. Gel strength (N mm) was determined by multiplying these two parameters.

Compression tests

For the texture profile analysis, samples (diameter, 25 mm and height, 25 mm) were compressed on the flat plate of the Instron texturometer with a cylindrical plunger (50-mm diameter) adapted to a 1000 N load cell at a deformation rate of 50 mm/min. Furthermore, sensitivity of the force measurement was 0.1 N. Based on preliminary trials to establish a compression limit that would ensure no cracking and recoverability of most samples, it was decided to compress samples to 60% of their height. In the test, each sample was compressed twice. The following parameters were determined: hardness (N), maximum height of first peak on first compression (in terms of eating quality, food's resistance at first bite); cohesiveness (A₂/A₁), ratio of second compression to first compression positive areas (maintenance of food resistance during chew down); gumminess (N), product of hardness and cohesiveness (strength required in the chew down process); springiness (L₂/L₁), ratio of the detected height of the product on the second compression to the original compression distance (ability of food to reacquire its initial shape and size after a first bite); chewiness (N), product of gumminess and springiness (albeit expressed in N, a measure of the energy spent in the chew down process).

For the compression-relaxation test, the compression procedure was as for the texture profile analysis except that the sample was compressed only once for one minute and the force exerted on the sample was recorded. Relaxation (%) was calculated as Y_T = 100 × (F₀-F₁)/F₀, where F₀ is the force registered at the onset of relaxation immediately after sample compression; and F₁ is the force registered after one minute of relaxation. Thus, (100-Y_T) is taken as an index of elasticity and is expressed as the percentage elasticity of the gel.

Sensory Analysis

Sensory evaluation was conducted by five panellists from INIAP/IPIMAR, extensively trained with the sensory scheme for sausage evaluation and capable of conducting a structured scaling of products.[24] Panellists participated in preliminary trials and in the experiments that led to the development of cod sausages. Moreover, panellists used other fish (hake) sausages as a reference.[6] Sausages were taken out from the package, tempered to about 20°C and cut into 2-cm-long slices. These slices were distributed in white plates and presented to the panellist in random order for evaluation. Mineral water was supplied to the panellists for rinsing between samples. The evaluation was performed in a room specifically conceived for sensory analysis and with adequate lighting. Panellists were asked to score several sensory parameters of the product, using a 0–5 scale: colour (0 – light to 5 – dark); cod and smoke odour (0 – absent to 5 - excessive); elasticity (0 – plastic to 5 – elastic), hardness (0 – soft to 5 – hard), succulence (0 – dry to 5 – succulent) and fat mouthfeel (0 – slightly oily to 5 – very oily); saltiness, cod and smoke flavour (0 – absent to 5 – excessive); and smoke and saltiness aftertaste (0 – absent to 5 - excessive).

Statistical Analysis

Chemical composition, pH, colour, and folding test determinations were performed in duplicate, the gel strength results are the average of six measurements and all other determinations were done in triplicate. A general linear model—one-way ANOVA—was used to determine significant differences (p ≤ 0.05) among sausages with different formulation. Multiple comparisons were done by the Tukey HSD test. All statistical treatment was done with the software STATISTICA^© from StatSoft, Inc. (Tulsa, OK, USA), version 6.1, 2003.

RESULTS AND DISCUSSION

Effect of Pea Protein and Carrageenan

Chemical composition

The chemical composition of cod sausages is presented in Table 2. The incorporation of 2 g pea protein/100 g sausage significantly (p < 0.05) increased protein and reduced moisture content. On the other hand, addition of 1 g carrageenan/100 g sausage only affected (p < 0.05) carbohydrate content, which increased from 6.9 ± 0.2 to 8.0 ± 0.2%.

Table 2 Chemical composition of cod sausages containing different levels of pea protein, carrageenan, and pork meat∗.

Pea protein level (g/100 g sausage)	Carrageenan level (g/100 g sausage)	Moisture (%)	Protein (%)	Fat (%)	Ash (%)	Carbohydrate (%)
0	0	72.2 ± 0.2^a	11.0 ± 0.1^a	7.0 ± 0.0^a	2.9 ± 0.0^a	6.9 ± 0.2^a
2	0	70.5 ± 0.0^b	12.4 ± 0.0^b	7.0 ± 0.3^a	3.0 ± 0.0^a	7.1 ± 0.3^ab
2	1	70.2 ± 0.0^b	12.2 ± 0.2^b	6.6 ± 0.0^a	3.0 ± 0.0^a	8.0 ± 0.2^b
Replacement by pork meat (%)
0		70.2 ± 0.0^a	11.7 ± 0.1^a	6.2 ± 0.6^a	1.8 ± 0.0^a	10.1 ± 0.6^a
25		69.8 ± 0.2^a	11.9 ± 0.1^a	7.1 ± 0.0^a	1.8 ± 0.1^a	9.5 ± 0.0^a
50		69.3 ± 0.1^b	11.5 ± 0.1^a	7.3 ± 0.7^a	1.8 ± 0.2^a	10.1 ± 0.9^a
∗Presented values correspond to mean ± standard deviation. Means within a column with different letters are significantly different (p ≤ 0.05).

pH and colour

Pea protein and carrageenan incorporation had no effect on pH values (Table 3). On the other hand, pea protein addition changed (p < 0.05) cod sausages colour towards a more yellow hue (higher b∗) and, hence, towards a higher chroma value (Table 3). Furthermore, a∗ showed a small increase (p < 0.05) with carrageenan addition, thus, meaning slightly redder sausages. The pea protein effect is related to the considerable level of incorporation and to ingredient's cream colour. Regarding carrageenan, a similar effect on colour was already observed in a previous study.[25]

Table 3 pH and colour parameters of cod sausages containing different levels of pea protein, carrageenan, and pork meat∗.

PEA protein level (g/100 g sausage)	Carrageenan level (g/100 g sausage)	pH	a∗	b∗	Chroma	Whiteness
0	0	6.21 ± 0.03^a	−0.28 ± 0.02^a	6.74 ± 0.03^a	6.75 ± 0.03^a	73.6 ± 0.2^a
2	0	6.21 ± 0.01^a	−0.29 ± 0.01^a	7.47 ± 0.02^b	7.48 ± 0.02^b	72.3 ± 0.3^a
2	1	6.19 ± 0.04^a	−0.14 ± 0.03^b	7.55 ± 0.11^b	7.55 ± 0.11^b	74.0 ± 0.2^a
Replacement by pork meat (%)
0		6.52 ± 0.04^a	−0.26 ± 0.09^a	6.24 ± 0.71^a	6.25 ± 0.70^a	70.8 ± 1.9^a
25		6.46 ± 0.00^a	−0.08 ± 0.00^a	6.29 ± 0.06^a	6.29 ± 0.06^a	70.8 ± 0.0^a
50		6.48 ± 0.01^a	−0.09 ± 0.07^a	6.89 ± 0.15^a	6.89 ± 0.15^a	70.8 ± 0.3^a
∗Presented values correspond to mean ± standard deviation. Means within a column with different letters are significantly different (p ≤ 0.05).

Instrumental texture

The results attained from the measurement of the textural properties are shown in Table 4. Pea protein addition had no significant effect on FT, elasticity, and cohesiveness. However, gel strength, hardness, gumminess, springiness and chewiness increased (p < 0.05) with pea protein incorporation. Moreover, carrageenan addition to sausages already containing pea protein further increased (p < 0.05) these parameters. Concerning other properties, only elasticity exhibited a significant (p < 0.05) change with carrageenan addition, namely, it fell from 45.9 ± 0.4 to 38.9 ± 1.1%. Therefore, sausages became springier and harder with pea protein incorporation, whereas carrageenan made sausages harder and springier but less elastic. Gumminess and chewiness variations were a consequence of these textural changes, and an analysis of gel strength components (see Material and Methods, Texture measurements, puncture test) showed that breaking force—related to the hardness parameter—increase (from 0.9 ± 0.1 to 1.3 ± 0.1 N and, then, with carrageenan, product C, to 1.6 ± 0.0 N) was the main factor causing gel strength variation. These results could be due to the absence of a water content adjustment, causing moisture to decrease while protein and carbohydrate contents increased (Table 2). Effectively, a greater protein concentration in the matrix may entail a firmness increase of the products.[26] On the other hand, additives such as soy protein or k-carrageenan present a filler reinforcement effect and, regarding non-muscle protein, is admitted that it can contribute to build a protein network structure during heating.[27] For cod sausages, this effect may have been even more important because cod protein is highly denatured as a consequence of the salting treatment. In fact, the texture of poor quality fish products may be substantially improved with the addition of certain non-muscle proteins.[28] Furthermore, results are in accordance with various authors, which have found textural advantages in the combination of proteins from leguminous seeds with hydrocolloids,[13] being observed a synergistic effect.[12]

Table 4 Textural parameters of cod sausages containing different levels of pea protein, carrageenan, and pork meat.∗

Pea protein level (g/100 g sausage)	Carrageenan level (g/100 g sausage)	Folding test	Gel strength (N mm)	Elasticity^∗∗ (%)	Hardness (N)	Cohesiveness	Gumminess (N)	Springiness	Chewiness (N)
0	0	2.0 ± 0.0^a	6.5 ± 0.6^a	46.1 ± 1.1^a	7.5 ± 0.3^a	0.13 ± 0.02^a	0.9 ± 0.2^a	0.31 ± 0.01^a	0.3 ± 0.1^a
2	0	2.0 ± 0.0^a	9.5 ± 1.2^b	45.9 ± 0.4^a	9.6 ± 0.4^b	0.16 ± 0.01^a	1.5 ± 0.0^b	0.34 ± 0.01^b	0.5 ± 0.0^b
2	1	2.3 ± 0.6^a	11.6 ± 1.3^c	38.9 ± 1.1^b	18.7 ± 0.7^c	0.16 ± 0.01^a	2.9 ± 0.3^c	0.44 ± 0.01^c	1.3 ± 0.1^c
Replacement by pork meat (%)
0		2.7 ± 0.6^a	14.5 ± 2.0^a	42.1 ± 1.7^a	16.9 ± 1.0^a	0.14 ± 0.02^a	2.4 ± 0.5^a	0.32 ± 0.03^a	0.8 ± 0.3^a
25		2.7 ± 0.6^a	14.2 ± 2.1^a	43.5 ± 1.7^a	23.1 ± 2.5^b	0.26 ± 0.04^b	6.1 ± 0.4^b	0.52 ± 0.05^b	3.2 ± 0.4^b
50		2.3 ± 0.6^a	19.6 ± 4.8^b	49.7 ± 2.7^a	29.1 ± 1.5^c	0.31 ± 0.01^b	9.2 ± 0.5^c	0.59 ± 0.02^b	5.4 ± 0.3^c
∗Presented values correspond to mean ± standard deviation. ∗∗Elasticity was measured by the compression-relaxation test. Means within a column with different letters are significantly different (p ≤ 0.05).

Sensory analysis

The sensory evaluation parameters of cod sausages are presented in Table 5.

Table 5 Sensory scores of cod sausages containing different levels of pea protein, carrageenan, and pork meat.∗

Pea protein level (g/100 g sausage)	Carrageenan level (g/100 g sausage)	Colour	Cod odour	Smoke odour	Elasticity	Hardness	Succulence	Fat mouthfeel	Cod flavour	Smoke flavour	Saltiness	Smoke aftertaste	Saltiness aftertaste
0	0	1.2 ± 0.4^a	0.6 ± 1.3^a	2.2 ± 1.3^a	0.8 ± 1.3^a	0.4 ± 0.5^a	2.4 ± 1.1^a	1.0 ± 1.0^a	1.4 ± 1.7^a	2.0 ± 1.0^a	4.2 ± 0.8^a	1.8 ± 1.3^a	3.6 ± 1.1^a
2	0	1.6 ± 0.9^a	1.0 ± 1.4^a	2.8 ± 1.1^a	0.8 ± 1.1^a	1.2 ± 1.6^ab	1.8 ± 1.3^a	1.0 ± 1.4^a	0.6 ± 0.9^a	1.6 ± 0.9^a	3.2 ± 1.8^a	1.8 ± 0.8^a	2.8 ± 2.3^a
2	1	1.2 ± 0.4^a	0.6 ± 1.3^a	2.8 ± 1.3^a	1.4 ± 1.1^a	2.8 ± 1.5^b	2.0 ± 1.6^a	0.6 ± 0.9^a	0.4 ± 0.5^a	1.8 ± 0.8^a	2.8 ± 1.6^a	1.8 ± 1.5^a	2.6 ± 2.1^a
Replacement by pork meat (%)
0		0.6 ± 0.5^a	0.0 ± 0.0^a	3.0 ± 1.2^a	1.2 ± 1.3^a	0.8 ± 1.5^a	1.0 ± 1.4^a	0.8 ± 0.8^a	1.6 ± 1.1^a	2.2 ± 0.8^a	1.8 ± 1.3^a	2.0 ± 1.0^a	1.6 ± 1.3^a
25		1.6 ± 0.5^ab	0.0 ± 0.0^a	3.2 ± 0.8^a	1.4 ± 1.5^a	2.0 ± 1.4^ab	1.2 ± 1.3^a	0.8 ± 0.8^a	1.0 ± 0.7^a	2.0 ± 1.0^a	1.4 ± 1.5^a	2.0 ± 1.0^a	1.0 ± 1.2^a
50		2.2 ± 0.8^b	0.0 ± 0.0^a	3.0 ± 1.2^a	1.8 ± 1.5^a	3.0 ± 1.0^b	1.4 ± 1.3^a	1.2 ± 0.8^a	0.8 ± 1.3^a	1.8 ± 1.1^a	1.0 ± 0.7^a	1.4 ± 0.9^a	1.0 ± 0.7^a
∗Presented values correspond to mean ± standard deviation. Means within a column with different letters are significantly different (p ≤ 0.05).

Combination of 2 g pea protein and 1 g carrageenan per 100 g sausage produced harder (p < 0.05) sausages, an increase from 0.4 ± 0.5 to 2.8 ± 1.5. All other properties remained unchanged. Hardness increase agrees well with the evolution of textural hardness (Table 4 and Figure 1). Other studies with fish sausages have also shown that instrumental hardness correlates strongly with the sensory hardness and both attributes increase in magnitude with a higher carbohydrate (starch) content.[29] Panellists found the sausages too salty, thus, the study of sodium chloride reduction in the following experiment.

Figure 1 Correlation between cod sausages' sensory and instrumental hardness.

Effect of Sodium Chloride

Chemical composition

The reduction of sodium chloride content caused a moisture increase and an ash content decrease (p < 0.05), as can be seen in Table 6. Thus, results agree with formulations (Table 1).

Table 6 Chemical composition of cod sausages containing different levels of sodium chloride and sodium bicarbonate.∗

Sodium chloride level (g/100 g sausage)	Moisture (%)	Protein(%)	Fat(%)	Ash (%)	Carbohydrate(%)
2.3	70.4 ± 0.1^a	12.3 ± 0.1^a	6.3 ± 0.4^a	2.8 ± 0.1^a	8.1 ± 0.5^a
1.2	72.0 ± 0.3^b	12.1 ± 0.1^a	6.2 ± 0.2^a	1.5 ± 0.0^b	8.2 ± 0.5^a
Sodium bicarbonate level(g/100 g sausage)
0	70.3 ± 0.3^a	12.1 ± 0.1^a	6.2 ± 0.8^a	1.5 ± 0.0^a	9.9 ± 0.4^a
0.15	70.3 ± 0.1^a	12.0 ± 0.3^a	6.5 ± 0.6^a	1.6 ± 0.1^ab	9.7 ± 1.0^a
0.30	70.3 ± 0.2^a	11.9 ± 0.2^a	6.2 ± 0.3^a	1.7 ± 0.0^b	9.9 ± 0.2^a
∗Presented values correspond to mean ± standard deviation. Means within a column with different letters are significantly different (p ≤ 0.05).

pH and colour

There was no effect of salt level on the cod sausages pH and colour (Table 7). This is an important result since colour has great influence upon consumers' choices.

Table 7 pH and colour parameters of cod sausages containing different levels of sodium chloride and sodium bicarbonate.∗

Sodium chloride level (g/100 g sausage)	pH	a∗	b∗	Chroma	Whiteness
2.3	6.49 ± 0.01^a	−0.11 ± 0.20^a	6.86 ± 0.39^a	6.86 ± 0.38^a	73.3 ± 0.2^a
1.2	6.52 ± 0.01^a	−0.19 ± 0.20^a	6.54 ± 0.62^a	6.55 ± 0.62^a	72.7 ± 1.2^a
Sodium bicarbonate level(g/100 g sausage)
0	6.40 ± 0.04^a	−0.27 ± 0.02^a	6.79 ± 0.08^a	6.80 ± 0.08^a	72.5 ± 0.2^a
0.15	6.87 ± 0.02^b	−0.31 ± 0.02^a	6.98 ± 0.06^ab	6.98 ± 0.06^ab	73.4 ± 0.1^b
0.30	7.10 ± 0.03^c	−0.32 ± 0.08^a	7.29 ± 0.13^b	7.30 ± 0.14^b	74.7 ± 0.2^c
∗Presented values correspond to mean ± standard deviation. Means within a column with different letters are significantly different (p ≤ 0.05).

Instrumental texture

The textural measurements made on high and low salt cod sausages are presented in Table 8.

Table 8 Textural parameters of cod sausages containing different levels of sodium chloride and sodium bicarbonate.∗

Sodium chloride level (g/100 g sausage)	Folding test	Gel strength (N mm)	Elasticity^∗∗ (%)	Hardness (N)	Cohesiveness	Gumminess (N)	Springiness	Chewiness (N)
2.3	2.3 ± 0.6^a	14.3 ± 0.6^a	41.8 ± 1.6^a	18.3 ± 0.9^a	0.22 ± 0.02^a	3.9 ± 0.6^a	0.49 ± 0.04^a	1.9 ± 0.4^a
1.2	2.7 ± 0.6^a	19.9 ± 1.5^b	43.4 ± 2.8^a	29.9 ± 1.3^b	0.31 ± 0.04^b	9.4 ± 1.3^b	0.64 ± 0.05^b	6.0 ± 1.3^b
Sodium bicarbonate level(g/100 g sausage)
0	2.0 ± 0.0^a	15.5 ± 1.1^a	38.9 ± 3.7^a	17.6 ± 0.9^a	0.16 ± 0.05^a	2.7 ± 0.9^a	0.41 ± 0.05^a	1.1 ± 0.5^a
0.15	2.3 ± 0.6^a	11.8 ± 2.8^ab	39.0 ± 0.5^a	15.3 ± 0.2^b	0.10 ± 0.03^a	1.6 ± 0.6^a	0.37 ± 0.03^a	0.6 ± 0.2^a
0.30	2.7 ± 0.6^a	10.4 ± 3.6^b	40.7 ± 1.0^a	22.9 ± 0.9^c	0.27 ± 0.00^b	6.3 ± 0.2^b	0.54 ± 0.02^b	3.4 ± 0.1^b
∗Presented values correspond to mean ± standard deviation. ∗∗Elasticity was measured by the compression-relaxation test. Means within a column with different letters are significantly different (p ≤ 0.05).

Significant (p < 0.05) textural differences were found between the formulations—an increase of gel strength, hardness, cohesiveness, gumminess, springiness and chewiness with sodium chloride reduction was observed. Only FT and elasticity remained unchanged. The positive effect of salt reduction means that sodium chloride effect on cod sausages is different from its effect on other fish products. Namely, in these foods, salt helps to dissolve myofibrillar proteins in aqueous matrices and, thus, contributes to the formation of a gelled protein network.[27] Therefore, the absence of a negative effect suggests that dissolution of myofibrillar protein and subsequent gelling has little impact on the final textural quality of cod sausages. This can be related to a high degree of protein denaturation, which can be seen in the poorer textural quality of cod sausages when compared with hake sausages with similar formulation.[6] The presence of carrageenan in the specific matrix of cod sausage could offer an explanation. There are known difficulties regarding use of gums in beef sausages with a considerable amount of salt, 2–3 g/100 g sausage.[20] Among possible factors that could hinder the application of gums is the incompatibility between salt-soluble muscle proteins and many polysaccharides under high salt conditions.[12] The rheological properties of many polysaccharide gums are substantially altered by salt level.[30] Moreover, authors working on beef sausage[20] reported that an increase of salt level from 1 to 2.5 g/100 g decreased textural properties of sausages containing ı-carrageenan, while κ-carrageenan sausages exhibited an opposite effect (in our experiment, a mixture of ı- and κ-carrageenan was used). Furthermore, for sardine mince with added ı-carrageenan, it was found that low salt gels were harder but less elastic and cohesive than high salt gels.[31] Taking these studies into account, cod sausages seem to require further study, namely, regarding the action of sodium chloride in their matrix.

Sensory analysis

Most sensory parameters did not show any salt reduction effect (Table 9). Smoke, salt flavour and saltiness aftertaste were reduced (p < 0.05) in the low salt sausages, product E, whereas hardness increased (p < 0.05). Besides those sensory properties with predictable variation, the reduction of smoke flavour suggests that it also depends on sodium chloride level. Hence, as in meat products,[32] salt has probably a flavour enhancing effect in cod sausages. A study conducted with beef sausages has shown that besides saltiness also flavour intensity scores become higher as salt content increases.[19] The sausages' hardening with less salt agrees with the textural measurements (see discussion in Texture, as well Figure 1). Salt reduction did not alter important sensory properties (as odour, elasticity, or succulence) and, regarding the altered parameters, low salt sausages were considered better by the panellists. Therefore, it seems that the major problems of salt reduction were overcome in cod sausages.

Table 9 Sensory scores of cod sausages containing different levels of sodium chloride and sodium bicarbonate.∗

Sodium chloride level(g/100 g sausage)	Colour	Cod odour	Smoke odour	Elasticity	Hardness	Succulence	Fat mouthfeel	Cod flavour	Smoke flavour	Saltiness	Smoke aftertaste	Saltiness aftertaste
2.3	1.6 ± 0.9^a	0.2 ± 0.4^a	3.4 ± 0.9^a	1.0 ± 1.0^a	0.7 ± 0.6^a	2.0 ± 1.6^a	1.8 ± 1.5^a	0.8 ± 0.8^a	3.0 ± 0.0^a	2.8 ± 1.9^a	2.0 ± 1.2^a	3.5 ± 1.3^a
1.2	1.2 ± 0.8^a	0.2 ± 0.4^a	3.2 ± 0.8^a	1.6 ± 1.3^a	3.2 ± 1.3^b	2.0 ± 1.6^a	1.4 ± 1.7^a	1.0 ± 0.7^a	1.3 ± 0.5^b	0.8 ± 0.5^b	1.2 ± 1.1^a	0.3 ± 0.5^b
Sodium bicarbonate level(g/100 g sausage)
0	1.6 ± 0.5^a	0.2 ± 0.4^a	3.0 ± 0.7^a	1.8 ± 1.5^a	1.0 ± 0.8^a	1.3 ± 0.5^a	0.8 ± 0.4^a	0.6 ± 0.9^a	2.0 ± 0.8^a	0.8 ± 1.1^a	1.5 ± 0.6^a	0.4 ± 0.9^a
0.15	1.0 ± 0.0^a	0.2 ± 0.4^a	2.2 ± 1.1^ab	1.8 ± 1.3^a	1.0 ± 0.8^a	2.0 ± 1.4^a	0.8 ± 0.4^a	1.0 ± 1.2^a	0.8 ± 1.0^b	0.8 ± 1.1^a	0.6 ± 0.5^b	0.6 ± 0.9^a
0.30	1.2 ± 0.4^a	0.6 ± 0.9^a	1.4 ± 0.5^b	1.8 ± 2.0^a	3.3 ± 1.3^b	2.0 ± 1.4^a	0.6 ± 0.5^a	1.0 ± 1.7^a	0.3 ± 0.5^b	0.6 ± 0.9^a	0.0 ± 0.0^b	0.6 ± 0.9^a
∗Presented values correspond to mean ± standard deviation. Means within a column with different letters are significantly different (p ≤ 0.05).

Effect of Sodium Bicarbonate

Chemical composition

Sodium bicarbonate addition only changed (p < 0.05) the ash content of cod sausages (Table 6). Ash content of those sausages containing 0.3 g bicarbonate/100 g was higher, thus, agreeing with formulations (Table 1).

pH and colour

Sausages pH increased (p < 0.05) with addition of sodium bicarbonate, from 6.40 ± 0.04 to 7.10 ± 0.03 (Table 7). Concerning colour, only a∗ parameter remained unaltered, b∗ parameter increased (p < 0.05), meaning a more yellow hue, chroma followed b∗ evolution and whiteness became also significantly (p < 0.05) higher (Table 7). A 0.7 pH variation with 0.3 g sodium bicarbonate/100 g sausage agrees well with results reported by other authors, since, for beef sausages, a 0.4 g/100 g level was enough to increase pH from 6.0 to 7.0.[19] Regarding colour, changes demonstrate that a small pH increase can cause important chemical modifications in proteins and other molecules, thus, leading to different properties. Nevertheless, trained panellists did not detect these colour changes (see below, Sensory analysis).

Instrumental texture

The measured textural properties of cod sausages containing different levels of sodium bicarbonate are shown in Table 8. Bicarbonate addition did not affect FT and elasticity. On the other hand, pH increase led to significantly (p < 0.05) harder, springier and more cohesive sausages (gumminess and chewiness followed these trends). Only gel strength decreased with progressively larger amounts of bicarbonate. This unfavourable variation of gel strength can be explained through analysis of its components (see Material and Methods, Texture measurements, puncture test): breaking force remained constant at 1.8 N, while breaking deformation declined from 8.6 ± 0.2 to 5.8 ± 1.9 mm. Therefore, higher pH caused a lower breaking threshold when sausages were penetrated with a metal probe. However, other texture properties showed more positive effects, which agreed with various studies that reported textural benefits in response to moderate pH increases. For ı-carrageenan treated beef sausages, hardness and rupture force increased more than 50% with an adjustment of the batter pH from 5.7 to 6.2.[20] Similarly to these products, cod sausages were also treated with a carrageenan (ı + κ) and had a low fat content, below 10% (Table 6). It has been noted that interaction and gelling of a low ionic strength myofibril/gum composite system can be greatly affected by pH.[33] The variation of protein charges as pH increased from 6.4 to 7.1 (isoelectric point of cod muscle proteins is pH 5.5) probably enabled more balanced electrostatic interactions between proteins and carrageenan, thus, favouring the formation of more rigid gels in the products.[20]

Sensory analysis

The results shown in Table 9 indicate that two fundamental changes (p < 0.05) took place with increasing sodium bicarbonate levels: smoke perception (odour, flavour and aftertaste) decreased and hardness augmented. All other sensory properties were unchanged. Once more, sensory hardness variation was consistent with the Instron textural analysis (Table 8 and Figure 1). Although a hardening effect with higher pH has been reported on beef sausages, other authors working also on beef sausages observed a hardness decline due to more bound water at higher pH.[19] However, taking into account that these beef sausages did not contain any gum, such as carrageenan, it becomes likelier that interaction between muscle protein and gum leads to a different correlation between texture and pH. Regarding taste, cod flavour scores did not fall with increased pH and panellists did not notice any off flavour, even at 0.3 g sodium bicarbonate/100 g sausage. This differs from other authors' findings, which reported increased off-flavour intensity with addition of 0.25 g sodium bicarbonate/100 g Frankfurters (34). Therefore, it seems advisable to increase pH of carrageenan containing cod sausages to about 7.0.

Effect of Pork Meat Substitution

Chemical composition

Replacement of cod mince by pork meat, up to a substitution degree of 50%, only caused a slight moisture decrease (p < 0.05) from 70.2 ± 0.0 to 69.3 ± 0.1% (Table 2). This variation was due to the lower moisture content of pork meat.

pH and colour

Pork meat substitution had no effect on pH and colour parameters (Table 3). The absence of colour changes because of cod mince replacement is unexpected since pork meat is redder and darker than cod mince. Furthermore, a similar study with hake sausages[6] showed that a substitution degree of 50% increased parameter a∗ from −0.60 ± 0.07 (no pork sausage) to 0.47 ± 0.09 (50-50 pork-hake sausage), meaning redder products. Nevertheless, there is a small non-significant a∗ increase from -0.26 ± 0.09 (no pork sausage) to −0.09 ± 0.07 (50-50 pork-cod sausage) and panellists detected a darker colour on the 50-50 pork-cod sausage (see below Sensory analysis). However, this sausage was clearly less red than the 50-50 pork-hake sausage. A possible explanation for this is the absence of curing salt in the pork-cod sausage formulation. The nitric oxide present in the curing salt reacts with the heme (a pigment associated to myoglobin) iron, producing nitric oxide myoglobin and, after heating, nitrosylo-hemochrome, a stable pink pigment.[35,36] Therefore, in spite of equal levels of pork (pork meat has much more heme pigment than either cod or hake), only addition of curing salt, as happened with pork-hake sausage,[6] would produce a redder colour.

Instrumental texture

With exception of FT and elasticity, all texture properties increased (p < 0.05) with higher amounts of pork meat in the cod sausages (Table 4). Four fundamental properties showed better results with a 50% replacement of cod by pork meat: gel strength, hardness, springiness and cohesiveness. The latter is a good indicator of protein denaturation[37]; thus, its increase from 0.14 ± 0.02 to 0.31 ± 0.01 expresses a great textural improvement and points to the poor quality of salted cod protein. Moreover, analysis of the gel strength components (see Material and Methods, puncture test) presented two opposite variations: with increasing amounts of pork, breaking deformation declined from 8.1 ± 1.1 to 6.6 ± 1.7 mm and breaking force had a large increase from 1.8 ± 0.1 to 3.0 ± 0.2 N. Therefore, gel strength variation also expresses the hardening of sausages with pork meat substitution. Besides, these results are consistent with other experiments: more meat muscle favours a hardening of the products.[2,10] This is a consequence of higher collagen content in terrestrial animals.[21] Another reason could be the observed moisture decrease with growing levels of pork meat, from 70.2 ± 0.0 to 69.3 ± 0.1% (Table 2). In spite of the textural improvement achieved with a 50-50 pork-cod sausage, product K, this product is softer, less cohesive and springy than a 50-50 pork-hake sausage with a similar formulation.[6] Poor quality of salted cod protein explains the difference. An important aspect of this study was the considerable textural variability between batches, namely, texture of the control sausages in the sodium bicarbonate (product F) and pork meat replacement (product I) experiments was quite different from texture of the lower salt sausages in the sodium chloride experiment (product E), albeit formulation was the same. Such variability is due to differences in the quality of raw material, namely, differences in the storage conditions of frozen cod. In spite of this variability, for each experiment, clear effects were shown.

Sensory analysis

Only two sensory parameters were modified: colour and hardness (Table 5). 50-50 pork-cod sausages (product K) were significantly (p < 0.05) darker and harder than no pork cod sausages (product I). Regarding hardness, its variation agrees well with textural measurements (Table 4 and Figure 1). On the other hand, colour differences found by panellists were not detected with the colorimeter measurements (Table 3). In this case, human eye was more sensitive than the equipment. Panellists also found 50-50 pork-cod sausages rather acceptable. In fact, acceptability range seemed quite broad, making it difficult to define distinctive lower and upper bounds of the values of hardness, as was done in other studies.[4]

CONCLUSIONS

Salted cod sausages became more yellow, springier, and harder with pea protein incorporation. Furthermore, carrageenan addition made products redder, harder, springier but less elastic. Sodium chloride reduction produced harder, springier, more cohesive, but less salty and smoky sausages. Textural effects of sodium bicarbonate addition were similar, with the exception of a decrease in gel strength, moreover, sausages were made less acidic, more yellow and lighter and less smoky. Replacement by pork meat had a considerable textural effect, since sausages became harder, springier and more cohesive. Therefore, all tested changes, namely, addition of pea protein (2 g/100 g), carrageenan (1 g/100 g), sodium bicarbonate (0.3 g/100 g), pork meat (26.7 g/100 g) and reduction of salt (1.2 g/100 g) had largely favourable effects on the sausages properties, particularly, in their textural quality. Pork meat substitution (up to 50%) was not fundamental for the overall acceptability of Frankfurter cod sausages. The good correlations between instrumental and sensory texture within each experiment results were also interesting.

However, a global correlation with all data was not so good. This could be related to the subjectivity of sensory hardness: whereas, within each experiment, different formulations were assessed simultaneously, between experiments, sensory evaluation sessions were held in different months. Therefore, future works should try to improve panellists' objectivity.

ACKNOWLEDGMENTS

This research was funded by QCA III MARE-FEDER Project 22-05-01-FDR-00006 “Qualidade e Inovação em Produtos da Pesca”. Furthermore, the authors would like to thank Dipl. Eng. Cátia Silva (Induxtra De Suministros Portuguesa, Lda.), Dipl. Eng. Ana Teresa Ribeiro (Escola Superior Agrária de Santarém) and Dipl. Eng. Nuno Raposo (EFAL, Lda.) for their excellent technical support to the study, namely, providing the various ingredients used in the experiments.