Effect of Pressure on the Salting and Ripening Process of Anchovies (Engraulis anchoita)
BEATRIZ E. FILSINGER
ABSTRACT
The effect of pressure on the salting and ripening process of anchovies was studied. Salt and water were determined throughout the process to establish the dynamics of salt penetration. The ripening of anchovies was evaluated by means of the total ester index and sensory assessments. The best quality product corresponded to an intermediate pressure of 131.5 g/cm*. Salt penetration and ripening was intermediate between the other two samples. Higher pressure increased salt penetration and diminished ripening velocity. These products, however, did not possess the best sensory characteristics. Lower pressure changed the ripening process and yielded an over-ripe product.
INTRODUCTION
THE PRESERVATION of foods by salting is competitive between spoilage due to tissue autolysis and microbial action and salt penetration that prevents thesephenomena. Salting of anchovies can be divided into two stages:the first corresponds to the diffusion of salt into the fish and the elimination of water from it. The second stage, longer and slower than the first, is ripening which renders a product with tender consistency and the characteristic pleasant aroma and taste. During this stage the fish is stored under pressure. Due to the applied pressure a brine containing biological material is extruded and completely covers the fish. Baldrati et al. (1977) found that different pressuresproduce significant changesin water activity and in dry weight of anchovies (Engraulis encrasicholus).
The purpose of this work was to determine the dynamics of salt penetration, the ripening velocity and sensoryassessments of salted and maturatedanchovies, which were preparedunder different pressures,while keeping constantthe other variables, such as temperature, salt weight, kind of salt and raw material characteristics.
MATERIAL & METHODS
ANCHOVIES (Engraulis anchoita) were caught on the Argentinian platform in the south-western Atlantic Ocean from 37.4”s to 38.8”s during spring, when they arrived at the coast for spawning. After being caught, the fish were held without ice for 2 or 3 hr before they arrived at the laboratory where they were immediately immersed for one day in saturated brine at room temperature (18-22°C). The fish were then manually beheaded, partially gutted in the same operation and packed in 10 kg cans, 24 cm high and 21.5 cm diameter. To reproduce industrial conditions a collar was put around each container which contained 13 kg anchovies and 2 kg salt. A layer of salt was first put in the container then a layer of fish and so on until the container was filled with alternate layers of salt and fish finishing with a layer of salt. A piece of wood (1 cm high &d 21.5 cm diameter) was placed on the top layer of salt. By means of concrete stones on top of the wooden discs the fish were pressed. The pressure was calculated as the quotient of total weight on the disc by the surface area of the container. The cans were divided into three lots of 18 cans (A, B and C) with the following pressures: A, 65.8 g/cm*; B, 131.5 g/cm*; C,
Author Filsinger is affiliated with lnstituto National de Tecno-
logia Industrial, Centro de lnvestigaciones
de Tecnologia Pes-
quera (CITEP), Marcel0 1. de Alvear 1168, 7600 Mar del Plats,
Argentina.
197.4 g/cm2. Cans were stored for ripening in a room at 18-22°C for 300 days.
Anchovies for chemical analysis were selected from the middle of the can. Anchovies were filleted and passed through a home meat mincer. All chemical tests were done in duplicate. Determination of salt was performed according to the AOAC (1960) method based on titration with silver nitrate. Water was determined by weight lost after 24 hr at 105 f 1°C (Boeri et al., 1978). Apparent water activity values were calculated by means of the equation a,,, = 1.002 0.042m proposed by Lupin et al. (198 l), where m is the NaCl molality considered to be in true solution in the total water content of the product. Total ester index was determined by a modification of the AOAC method (Filsinger et al., 1982). Ten grams anchovies were ground in cooled distilled water in a Virtis homogenizer and made up to 100 ml. Aliquots of 20 mL were placed into distilling flasks. Into each flask was added ca. 1 ml phenolphthalein and the contents titrated to pink using 0.5 N KOH. An additional 25 mL 0.5N KOH was placed in each solution, the sample was refluxed for 1 hr, and the excess alkali was titrated with 0.1 N HCl. Since the measurements were done in aqueous homogenates instead of purified lipid extracts, the values obtained are representative of total ester indices. Results were expressed in grams KOH on a corrected dry weight (CDW) basis. Corrected dry weight was calculated by subtracting the salt content from the dry weight. In the figure where total ester index vs time is plotted, the slope of the straight line is the ripening velocity. Total ester index remained constant following ripening, i.e., it had a zero slope (Himmelblau, 1970). The intercept of both lines is the ripening time.
Sensory assessments
Sensory scores were assigned according to the scoring method presented in previous work (Filsinger et al., 1982). The attributes considered were flavor, flesh color, odor, flesh consistency and flesh adherence to the backbone. The anchovies were sensorily evaluated by a panel of six people experienced in judging fish quality. The judges were previously trained in the fish processing industry according to the current methods used to assess the degree of ripening of salted anchovy. A scale of points from 0 to 8 replaced the ambiguous “unripe” or “green”, “ripe” and “over-ripe” terms which are traditionally used by expert tasters in the fish processing industry. A fish is assigned a score for each attribute according to the descriptions in the table. The average of the five attributes was taken as the score for the fish. The final score given to the sample was the average of thirty six specimens evaluated by the six judges.
RESULTS & DISCUSSION
Salting process
Figure 1 shows the changesin salt content (x,), water content (x,) and apparentwater activity during salting for cans of lot B.
Figures 2 and 3 show the changesin salt and water, respectively, for the different lots. The fish with the highest pressure reachedequilibrium faster, attaining the highest salt and lowest water content.
The following mathematical model proposed by Zugarramurdi and Lupin (977) was usedto explain the salting process.
dx
dt
,=
k
s
K
-
xs)
dx,dt
-
kv
<XL - x,1
(2)
Volume 52, No. 4, 1987JOURNAL
OF FOOD SCIENCE-919
SALTED AND RIPENED ANCHOVIES. . .
20 LO
60
@a
100
120
TIME ( hours )
Fig. l-Changes
in salt, water and water activity during ripening
of anchovies. x- - -x, salt content; l - -e, water content; o- - -0,
water activity. COW-corrected dry weight calculated by subtract-
ing the salt content from weight obtained by drying.
TIME I hours 1
Fig. 2-Changes
in salt during ripening of anchovies. Pressure:
Lot A, 65.8 glcm2, -‘-‘-‘; Lot B, 131.5 glcm? - - - -; Lot C, 197.4
glcm2, -.
where Mass of salt at a given moment
x, = Total mass - mass of water - mass of salt (3)
Mass of water at a given moment x, = Total mass - mass of water - mass of salt
(4)
Total mass - massof water - mass of salt is the CDW; t is the time; k, and k, are the specific rate constantsfor the variations of salt and water, respectively. The “*” values are at equilibrium.
Eq. (1) and (2) can be easily integrated with the following initial conditions
(t=o): x,(O)= x: and x,(O) = x0\_
(5)
The solutions are:
x, = x: . emkst + x; (1 -e-ks’)
(6)
x, = x$ . e- kwt + x; (1 -,-kwt)
(7)
In order to study the correlation between experimental data and theoretical expression (6) and (7), the integral method of Swinboume (1960) and Mangelsdorf (1959) was used. This method basically consists in a transformation of exponential expressions into a linear one, using the time interval (T) between two consecutive experimental determinations. The following expression relates the salt concentration at time (t + T)
to the value at time t:
x,(t+T) = xi (1 -e-kst) + e-ksT. x&t)
(8)
A similar expression can be obtained for x,(t + T) in terms of
&v(t). According to Eq. (8) the plot of x,(t + T) against x,(t) is a
straight line, where the first term of the secondmember is the y-intercept and e- ksTis the slope. The valuesof x&t + T) against the corresponding x,(t) values for lot B are plotted in Fig. 4. From the slope of the linear it was possible to calculate k,. The values of k, and k, for the three lots calculated in a similar way are presentedin Table 1. The correlation coefficients and the 95% confidence limits are also given. The results showed
higher values of k, and k, when pressure increased. Moreover,
the highest increasesof dry weight (or lowest water) and the highest salt were found in the anchoviesfrom lot C. The fastest decreaseof apparentwater activity was observedfor lot C (Fig. 5). For equal amounts of added salt higher pressuresproduced lower apparent water activities.
Fig. 3-Changes
in water during ripening of anchovies. Pres-
sure: Lot A, 65.8 glcm2, .-.-.-; Lot B, 131.5 glcmz, - - - -; Lot C,
197.4 glcm? -.
z,I
91
0.2
0.3
0.4
Salt Concentration (gNaCl/gCDW) at time t
Fig, 4-Relation
of salt concentration
centration at time t for lot B.
at time, (t+ T), to salt con-
92O-JOURNAL OF FOOD SCIENCE-Volume 52, No. 4, 1987
0.75 1 0
I
I
I
I
20
40
60
80
I
I
100
120
TIME ( hours I
Fig. 5-Changes
in water activity during ripening of anchovies.
Pressure: Lot A, 65.8 glcm2, -.-.-; Lot B, 131.5 glcm2, - - -; Lot C,
197.4 glcmZ, -.
Table I-Values of the specific constants and confidence limits probability for pressure-processed anchovies
Salt
Water
Sample
k, (hr-‘)
‘
k, (hr-‘)
Lot Aa Lot Bb Lot c=
0.06732 k 0.008
0.99
0.08474 2 0.005
0.07318 t 0.010
0.97
0.09015 ? 0.007
0.08727 k 0.006
0.98
0.10337 ? 0.008
a 65.8 g/cd pressure
b 131.5 g/cm2 pressure c 197.4 g/cm2 pressure
for 95%
r 0.96 0.94 0.92
Ripening process
Figure 6 shows the results of total ester index vs time for lots A, B and C. The slope of the straight line is the ripening velocity. Total ester index remained constant during the storage after complete ripening. The intercept of both lines is the ripening time.
Figure 7 shows the results of sensory scores vs time. The ripening times evaluated by means of sensory scores approx-
0 0
I
I
I
I
I
100
160 200 233
Rlpenlng trme (clays?
Fig. 7-Sensory
scores of anchovies at various periods of rip-
ening. Each point is representative
of at least 36 samples and
is the average of the sensory characteristics
evaluated accord-
ing to the scoring method presented in previous work (Filsinger
et al., 1982). Pressure: Lot A, 65.8 glcm2, -.-.-; Lot B, 131.5 gl
cmZ,----.
I Lot C, 197.4 glcmZ, -.
Sensory scores: 0, raw
fish; 2, beginning of ripening; 4, the middle stage; 6, complete
ripening; and 8, deteriorated or over-ripe anchovy. Odd num-
bers were reserved for intermediate stages. alntercepts indicate
complete ripening time.
Table Z-Ripening velocity and time of ripening for pressure processed anchovies
Sample
Lot Ab Lot BC Lot Cd
Ripening velocity by TEI method
(gKOH/lOOga day)
0.06726 0.03757 0.02884
Time of ripening by TEI method (days)
178 223 >254
Time of ripening by sensory scores
(days)
160 233 >280
a On corrected dry weight basis
b 65.8 g/cm2 pressure
c 131.5 g/cl+ pressure d 197.4 glcm2 pressure
imately agreed with those determined by means of the total ester index.
Ripening velocity and ripening time for the three lots are presented in Table 2. The results showed that the ripening process was slower when pressure was increased.
Anchovies of lot C never reached ripening, so flesh completely lacked elasticity, and the backbone adheredvery tightly to the flesh which was tom in the filleting process. There was no uniformity in color which showed dark red blots. Very little pressure (lot A) yielded an over-ripe product which was characterized by: flimsy flesh which tore in the filleting process; color was uniform and odor was lightly rancid. The best quality anchovies obtained in this work were those of lot B, where color was uniform, flesh was firm and resistant and separated neatly from the backbone with the characteristic flavor of ripened anchovies.
Ripeniog time (days 1
Fig. GChanges
in Total Ester Index during ripening of an-
chovies. Pressure: Lot A, 65.8 glcm2, -.-.-; Lot B, 131.5 glcmz, -
\_ \_ \_I. Lot C, 197.4 glcmZ, -.a
Intercepts indicate complete
ripening time.
CONCLUSION
HIGHEST PRESSURE lowered the ripening velocity but ac-
celerated the penetration of salt. This faster salt penetration
decreasedwater activity and inhibited bacterial growth more
rapidly, but this product did not possessthe best sensory char-
acteristics. Very little pressure changed the maturation and
yielded an over-ripe product.
-Continued on page 927
Volume 52, No. 4, 1987JOURNAL OF FOOD SCIENCE-921
Ver+/-
