In Cheese Making Why Can You Never Use Iodized Salt

Introduction

Iodised salt, used as an additive in food production, is the principal source of iodine in Switzerland (Haldimann et al. 2015). Salt iodisation was first introduced in Switzerland in 1922 at a level of 3.75 mg kg⁻¹. In the years to follow, the iodine concentration in salt has been increased gradually to the current value of 25 mg kg⁻¹. However, the iodine concentration in the salt cannot be increased further to any degree, as this can lead to an excessive intake of iodine in individuals with a high salt consumption. The WHO recommends iodine fortification levels in the range of 20–40 mg kg⁻¹. To reduce the risk of iodine deficiency, a high penetration rate of iodised salt in processed food is important to achieve an iodine supply that meets the nutritional requirements of different population groups.

To ensure an adequate iodine supply, the WHO recommends a minimum urinary iodine concentration (UIC) threshold of 100 µg l⁻¹ for infants, children and adults, and 150 µg l⁻¹ for pregnant women. Despite the increasingly higher concentrations of iodine in salt, iodine supply of the Swiss population appears inadequate as the average UIC level is lower than recommended by WHO (~100 µg l⁻¹), while the risk of iodine deficiency is higher in certain groups, including adults (76 µg l⁻¹), pregnant (140 µg l⁻¹) and breast-feeding women (75 µg l⁻¹), infants (91 µg l⁻¹) and children up to 6 months (91 µg l⁻¹) (Andersson and Herter-Aeberli 2018; Andersson et al. 2019).

Milk and dairy products are generally good dietary sources of iodine (Haldimann et al. 2005; Bath et al. 2012); however, they are prone to variations in iodine concentration. Large fluctuations in iodine content of milk and dairy products originate mainly from supplements to cattle feed (Schöne et al. 2017; van der Reijden et al. 2018) and seasonal variation (Walther et al. 2018). During cheese production, a large part of milk iodine is lost because, like other substances dissolved in milk serum, iodine passes into the whey, which is separated from the curd. The use of iodised salt in brine treatment and cheese curing could improve the contribution of cheese to the iodine supply in Switzerland, where the per capita consumption of cheese, at approximately 21.3 kg, is high (TSM Treuhand GmbH 2017).

Salt contributes directly to the taste of cheese and influences the growth of bacteria, both on the cheese surface and in the cheese body. For the salting of cheese, various processes such as curd salting (e.g. cheddar), surface salting (e.g. soft cheeses), salting in brine and brushing with salt water (e.g. smear-ripened cheeses) are often used in combination. During immersion in brine, the aqueous phase diffuses into the cheese due to osmotic pressure differences and a salt gradient develops from the surface to the centre (Guinee 2004). Most studies described the diffusion of salt in cheese during brining and ripening using Fick's law (Tyrrell 1964; Guinee and Fox 1983). In cheese, the moisture is kept in a protein matrix that also incorporates fat globules. Diffusion of salt across a cheese loaf may be regarded as a transfer mechanism in a porous medium, where the pores are connected and filled with an aqueous phase. As the pores are not aligned, the diffusion takes place over a longer distance than it would in homogenous material (Simal et al. 2001). Sodium and chloride are associating solutes, i.e. the two ions are linked together electrostatically, and thus diffusion of both ions is limited by the slower ion. In addition to the size of the moving particles, various cheese properties such as porosity, tortuosity, moisture (on a fat free-basis) and viscosity of aqueous phase as well as chemical interaction with other charged components in both the aqueous and the solid phases influence the rate of salt diffusion. It is important to note that the retention of sodium as a result of ion exchange with calcium does not take place in casein micelles (Floury et al. 2009).

The diffusion coefficient (D) is a proportionality constant that is used to describe the mass transport of a substance through homogeneous material using Fick's law (Tyrrell 1964). In contrast, the apparent diffusion coefficient (D_app) takes into account not only the random motion of the particles (Brownian molecular motion), but also matrix-dependent factors such as porosity and sorption effects. D_app is a measure of the mobility of the particles that can be determined from the distance that the particles have migrated through a medium in a given time. Hence, the experimental determination of D_app allows comparison of the rates of diffusion of different substances in a matrix. The apparent diffusion coefficient for NaCl in cheese moisture was reported to be typically 2.3 × 10⁻¹⁰ m²s⁻¹, although it varies from 1.2 × 10⁻¹⁰ to 5.2 × 10⁻¹⁰ m²s⁻¹ according to cheese composition and brining conditions (Geurts 1974; Guinee 2004). In contrast, the diffusion coefficient (D) of NaCl in pure water at 12.5°C is 1.2 × 10⁻⁹ m²s⁻¹ (Friedman and Kennedy 1955). Moreover, in a comparative study, it was demonstrated that sodium and potassium cations diffused at a similar rate from the brine into cheese (Reps et al. 1995).

Few studies explored the use of iodised brine for iodine enrichment of cheese. Potassium iodate (KIO₃) has been used to study the iodine transfer from brine (20 mg kg⁻¹ iodine in salt) to Edam cheese (Wiechen and Hoffmann 1994). The diffusion of iodate (IO₃ ⁻) was monitored by measuring a radioactive iodine-131 tracer. The study revealed vast differences between the iodine content of the 1 cm thick border zone (399 µg kg⁻¹) and the core (52 µg kg⁻¹) after a ripening period of 42 days. Presumably, iodate was reduced to elemental iodine, which was retained in the cheese rind as a result of iodine-protein interaction. Thus, the authors did not recommend the use of iodised brine.

Similarly, a study with Feta cheese showed that iodate transport is a complex process, which cannot be explained by a constant diffusion coefficient; a part of the iodine is trapped in the protein matrix (Vosniakos et al. 1992). In contrast, the iodination of brine (18.5 mg kg⁻¹ iodine in salt) with KIO₃ was found to be an effective measure to increase the iodine content of semi-hard cheese (Hoffmann et al. 1997). Also, dry surface salting resulted in an increased iodate content in Camembert soft-cheese (Zimmermann et al. 2005). In Switzerland, salt used for iodine supplementation is iodised with potassium iodide (KI). For this reason, it is not possible to apply the results of the studies mentioned above to estimate how much iodine can be absorbed by cheese.

As a halogen element, iodide should behave similarly to chloride, but because of the size and atomic mass, the iodide ion (I⁻) is likely to diffuse more quickly than iodate (IO₃ ⁻). In contrast to iodate, there is no information on the migration behaviour of iodide in cheese. For a better understanding of the dissemination of iodised salt into cheese, knowledge of the diffusion behaviour of iodide in different types of cheese is required. Therefore, the main objective of this work was to study the diffusive migration of iodide into different cheese types after immersion in brine and subsequent ripening. A complementary analysis of chloride allowed a direct comparison of the migration behaviour of the two halogenide ions. Based on these results, an improved estimate of the potential iodine supply from cheese for the Swiss population may be achieved.

Materials and methods

Production of experimental cheeses

Four hard cheeses, four semi-hard cheeses and a batch of soft cheeses were produced in an experimental cheese plant (Agroscope, Liebefeld, Switzerland).

The hard cheeses (Gruyère-type, diameter 30 cm, weight 8.3 kg) were produced from 90 l of raw full-fat (37 g kg⁻¹) cow milk. After the addition of 70 ml of the starter cultures RMK 101 and RMK 124 (Agroscope, Liebefeld, Switzerland), which consist of strains of Lactobacillus delbrueckii subsp. lactis, and Streptococcus thermophilus, the milk was pre-ripened at 31–32°C for 15 min. For coagulation, 15 ml of rennet was diluted in 1 l of water and added to the milk, which was then held at 32°C for 35 min. The rennet (Winkler GR orange) was obtained from Winkler AG (Konolfingen, Switzerland). Its strength is described as follows: 1 part of the rennet clots 9000 parts of non-heated full fat cow milk (pH 6.65 at 32°C) within 30 min, which is equivalent to 194 international milk clotting units (IMCU) ml⁻¹. The coagulum was cut into grains of about 3–6 mm size using a cheese harp with vertical wires. Afterwards, the curds–whey mixture was warmed to 56°C for 30 min followed by a final stirring (56°C, 10 min). To remove the whey, the curds were transferred into perforated moulds (30 cm) and pressed for 4 h. The cheeses were then immersed in brine solution 20% for 16 h at 11–13°C and finally ripened at 14–15°C and 90–96% relative humidity for 180 d. During the first 10 d of ripening, the cheeses were smeared daily with brine solution 4% which had been previously inoculated with a mixture of Brevibacterium linens, Arthrobacter ssp. and Debaryomyces hansenii (OMK 702, Agroscope, Liebefeld, Switzerland); after that, the brine solution was applied twice a week.

The semi-hard cheeses (Tilsit-type, diameter 30 cm, weight 7.2 kg) were produced from 70 l of pasteurised full-fat (37 g kg⁻¹) cow milk with addition of 5.0 ml calcium chloride 35%. After the addition of 5 l of water and starter cultures (30 ml each of MK 420 and RMK 150; Agroscope, Liebefeld, Switzerland) that consist of strains of Lactobacillus delbrueckii subsp. lactis, Streptococcus thermophilus, and Lactococcus lactis subsp. lactis, the milk was pre-ripened at 31–32°C for 15 min. For coagulation, 10 ml of rennet (Winkler GR orange) was diluted in 1 L of water and added to the milk, which was then incubated at 32°C for 35 min. The coagulum was cut into cubes of about 10 mm size using a cheese harp with vertical wires. Thereafter, 20 L of water was added to the curds–whey mixture which was warmed up to 44°C for 20 min followed by a final stirring (43°C, 20 min). To remove the whey, the curds were transferred into perforated moulds (30 cm) and pressed for 7.5 h. Cheeses were then immersed in 20% brine solution for 14 h at 11–13°C and finally ripened at 14–15°C and 90–96% relative humidity for 90 d. Cheese-curing was made in the same way as described above for the Gruyère-type hard cheeses.

The soft cheeses (Camembert-type, diameter 12 cm, weight 0.4 kg) were produced from 26 l of pasteurised full-fat (37 g kg⁻¹) cow milk with the addition of 5.0 ml calcium chloride 35%. After the addition of the starter cultures RSW 901 (5‰, Lactococcus lactis ssp. lactis, Lactococcus lactis ssp. cremoris and Lactococcus lactis ssp. diacetylactis) and Yoghurt B1 (5‰, Lactobacillus delbrueckii ssp. bulgaricus, Streptococcus thermophilus) (Agroscope, Liebefeld) and the surface cultures OMK 701 (Geotrichum candidum; inoculation level 1‰) (Agroscope, Liebefeld) and CHOOZIT PC 22 LYO (Penicillum candidum, inoculation level 1‰ 10⁴ cfu ml⁻¹) (Joh. Bichsel AG, Grosshöchstetten, Switzerland), the milk was pre-ripened at 35°C for 40 min in a climate chamber. For coagulation, 6.5 mL of rennet (Winkler GR orange) was diluted in 0.5 l of water and added to the milk, which was then held at 35°C for 30 min. The coagulum was cut into cubes of about 30–40 mm size using a small cheese harp with vertical wires. After that, 5 l of water (35°C) was added to the curds–whey mixture which was held at 35°C for 30 min and gently turned over twice. The curd-whey mixture was filled into perforated moulds (12 cm) which were turned over after 10 min, 1 h and 4 h during draining (20 h, 25°C). For salting, the cheeses were immersed in brine solution 20% for 30 min at 11–13°C and then the cheeses were allowed to dry for 2 days (13–15°C, 80 % R.H.) and further ripened (14–15°C and 90–96% R.H.) for 12 d.

To study the influence of iodised and non-iodised salt on the iodine concentration in consumers cheese, half of the produced cheeses were brine-salted and smeared using saline solutions prepared with iodised salt, whereas for the other half of the cheeses saline solutions prepared with non-iodised salt were used. The brines containing iodised or non-iodised salt were made from 180 l hot water (70°C) by addition of 20% sodium chloride (Schweizer Rheinsalinen, Pratteln, Switzerland). According to the specifications of the manufacturer, the iodised salt was supplemented with 20 mg iodine per kg in the form of potassium iodide (KI). Calcium chloride was added to obtain a final concentration of 0.22%. Subsequently, the pH-value was adjusted to 5.2 by the addition of lactic acid 90%. (Dr. Grogg Chemie AG, Bern, Switzerland). The cheeses were smeared with a 4% saline solution containing iodised or non-iodised salt, respectively and a surface culture.

Sampling

Loaves of Camembert-type cheeses were sampled after 14 days of ripening and stored at −20°C until analysis. Samples of Tilsit-type semi-hard cheeses were taken after 45 days and 90 days of ripening. Samples of Gruyère-type hard cheeses were sampled after 45 days, 90 days and 180 days of ripening. At each sampling of the experimental Tilsit and Gruyère cheeses, three vertical cylinders of 2.5 cm diameter were removed from the loaf at a distance of half of the radius.

As depicted in Figure 1, two of the sampled vertical cylinders were cut into five pieces to carry out zonal analyses (rind, border and core zone) of the salt and iodine contents at different stages of ripening. In a first step, the upper and lower cheese rind was cut away (width 0.5 cm), then the upper and lower border zone (width 2.5 cm) were cut away to obtain the core zone (width 4 cm). The upper and lower cheese rind, as well as the upper and lower border zone, were pooled.

Increase of iodine content in brine-salted soft, semi-hard and hard cheeses by diffusion of iodide

Published online:

25 September 2019

Figure 1. Sampling scheme during cheese ripening. Three cylinders each were cut out of the semi-hard cheeses (Tilsit-type) after 45 and 90 days and additionally after 180 days out of the hard-cheeses (Gruyere-type). Each time, the rind of one of the cylinders was removed to obtain a sample corresponding to the edible part of the cheese. The two remaining cylinders were cut into zonal samples (rind, border and core zone) as illustrated.

Figure 1. Sampling scheme during cheese ripening. Three cylinders each were cut out of the semi-hard cheeses (Tilsit-type) after 45 and 90 days and additionally after 180 days out of the hard-cheeses (Gruyere-type). Each time, the rind of one of the cylinders was removed to obtain a sample corresponding to the edible part of the cheese. The two remaining cylinders were cut into zonal samples (rind, border and core zone) as illustrated.

The third cylinder was cut into 3 pieces as follows: the upper and lower cheese rind were cut away (width 0.5 cm) and pooled with the rind parts of the two other cylinders to have enough material for the analysis. The rindless cylinder was used for the determination of the average iodine and salt content in the edible part of the cheese.

At the end of ripening, a v-shaped sample of about 300 g was cut away from each cheese loaf for the analysis of cheese composition. The rind was removed, the cheese samples grated and the sample material further homogenised by shaking the grated cheese in a plastic bag. In addition to the cheese samples, samples of vat milk, whey and brine were collected for analysis of iodine content. All samples were immediately frozen at −20°C until further analysis. Before analysis, the samples were allowed to defrost in a fridge overnight.

Analysis of cheese composition

The water content was determined by weight difference using the IDF reference method for the determination of the total solids content (ISO 2004). The fat content of the cheese was determined by the Gerber-van Gulik method (ISO 2008a). The nitrogen content and crude protein was determined by the Kjeldahl method (ISO 2008b). The sodium chloride concentration was obtained using the potentiometric titration method (ISO 2006).

Determination of iodine content in cheese

For the sample preparation, the cylindrical cheese samples were cut into smaller pieces and weighed in graduated 50 ml PP-tubes (Sarstedt, Nümbrecht, Germany). Then, 5 ml (or 10 ml for larger samples) of 25% tetramethyl ammonium hydroxide (TMAH; Trace Select, Sigma-Aldrich, Buchs, Switzerland) were added to each tube and diluted with water to approximately 45 ml. After mixing, the tubes were closed with screw caps and put in a heat cabinet for 4 h at 90°C for solubilisation and iodine extraction. The sample tubes were shaken regularly during the extraction period. After cooling to ambient temperature, the tubes were filled with water to 50 ml. The suspensions were left to settle for at least one hour. To avoid contamination with sediment particles and the fat layer, 0.5 ml aliquots of the extracts were pipetted from the middle part of the tubes into graduated 12 ml PP-tubes (Semadeni, Ostermundigen, Switzerland) and diluted with 9.5 ml 0.01% TMAH. Then, the solution was spiked with 25 μl 1.34 μg/ml iodine-129 isotope (¹²⁹I) standard, which was prepared from the NIST SRM 4949C radioactivity standard (Standard Reference Material 4949C, National Institute of Standards and Technology, Gaithersburg, MD, USA). Sector-field inductively coupled mass spectrometry (ICP-MS, Element XR, Thermo, Bremen, Germany) was used to determine the iodine concentrations in cheese and brine.

Modelling of apparent diffusion coefficients in cheeses

Fick's law describes the diffusion flux through a porous matrix such as cheese in a one-dimensional system. Assuming a non-steady-state with constant diffusivity, Fick's diffusion equation describing mass transfer can be written as, (1) $\frac{\mathrm{\partial }C}{\mathrm{\partial }t}=D\frac{{\mathrm{\partial }}^{2}C}{\mathrm{\partial }{x}^2}$ (1)

where t is the time (s) of diffusion, C is the concentration (μg kg^{- 1}) of the anion iodide or chloride in the aqueous phase of the model cheese matrix, and D is the apparent diffusion coefficient (m²s⁻¹) for iodide or chloride in the cheese matrix. Equation (1) describes the change of the molar concentration of anions under study in the aqueous phase of cheese with respect to time at a given distance from the surface. For studying iodide and chloride diffusion in semi-hard (Tilsit) and hard cheese (Gruyère), it was assumed that the mass transfer is unidirectional along the vertical axis, i.e. the diffusion iodide from brine and smear salt treatments enters predominantly through the flat sides of the wheel-shaped cheese loaf, with negligible quantities through the hoop side. Using this simplification, the variable x (m) is the position along the vertical axis of the cheese. Initially, the surfaces at the cheese-brine interface are kept at a uniform concentration (C_i). Thus, the initial and boundary conditions are as follows:

t = 0 → C (x, t) = C_i 0 < x < L

L is the height of the cheese wheel. When M_t is the total quantity of diffusing iodide (or chloride) that has entered the cheese at time t, and M_∞ is the corresponding quantity after infinite time, the solution of Equation (1) takes the form: (2) $\frac{M_t}{M_{\mathrm{\infty }}}=1-\frac{8}{{\pi }^2}\sum _{n=0}^{\mathrm{\infty }}\frac{1}{{\left(2n+1\right)}^2}.exp\left\{-\frac{{\left(2n+1\right)}^2{\pi }^2}{L^2}Dt\right\}$ (2)

M_t was calculated from measured concentrations in the sample cylinders (Figure 1) at different ripening times t. M_t at the time t = 0 reflects the quantity of native iodine from milk, which was obtained from the comparative experiment carried out without iodised salt. M_∞ was estimated as the amount of diffusing iodine at the end of the ripening time by fitting the best profile to the measured concentrations at different positions (L) in the cheese. Hence, the iodine uptake at L = 0 was considered an equilibrium concentration. The diffusion coefficient D was taken from Equation (2) after iteratively minimising the sum of squares of the deviations (S) between the experimental and theoretical M_t/M_∞ ratios according to the model: (3) $S=\sum _{i=1}^n\left(\left.{\left(\frac{M_t}{M_{\mathrm{\infty }}}\right)}_{exp.}-{\left(\frac{M_t}{M_{\mathrm{\infty }}}\right)}_{model}\right)\right.$ (3)

If the dispersion between experimental and theoretical data is weak, the mathematical model fits the experimental data. In the following text, we refer to the apparent diffusion coefficient as D_app.

Results and discussion

Chemical composition of the experimental cheeses

The composition data of the experimental Camembert, Tilsit and Gruyère-type cheese are shown in Table 1. The firmness of cheeses is characterised by the moisture on a fat-free basis (MFFB) that is expected to be > 670 g kg⁻¹ for soft, 540–690 g kg⁻¹ for semi-hard and 490–560 g kg⁻¹ hard cheese (Codex 1978). During smear ripening of cheeses, a continuous weight loss in the range of 10% of the initial weight is normal due to the evaporation of water at the cheese surface. As a result of the continuous water evaporation, smear-ripened semi-hard cheeses may shift to the category of hard cheeses towards the end of ripening. The experimental Tilsit-type cheeses at the end of ripening showed a MFFB of 531 g kg⁻¹, which is slightly below the required value for semi-soft cheeses (540–690 g kg⁻¹). This slight deviation can be explained by the increased moisture loss of the cheeses in the ripening cellar of the experimental cheese plant that was not utilised to full capacity as in practice. Similarly, the Gruyère-type cheeses, which were ripened in the same cellar, had an average MFFB of 495 g kg⁻¹ after a ripening period of 180 d which is in the lower range of the scale for hard cheeses. For commercial samples of Swiss Tilsit and Gruyère cheese, MFFB values of 538 ± 11 g kg⁻¹ and 515 ± 21 g kg⁻¹ and FDM values of 493 ± 20 g kg⁻¹ and 511 ± 15 g kg⁻¹ have been reported previously (Bütikofer et al. 2008). In a recent study (unpublished results, 2010) we found an average salt content of 17.2 ± 3.3 g kg⁻¹in 15 samples of Camembert from Swiss retailers. In the experimental Camembert cheese produced for the present study, the salt content was lower (13.0 ± 0.7 g kg⁻¹). In contrast, the average salt contents of the experimental Tilsit (21.3 ± 1.1 g kg⁻¹) and Gruyère (18.3 ± 0.9 g kg⁻¹) cheeses were higher than the reference values reported for these types of cheese (17.5 ± 2.0 and 15 ± 2.3 g kg⁻¹, respectively) (Sieber 2011). Even though the chemical composition of the experimental cheeses deviated slightly from reference values, the overall quality of the cheeses produced was satisfactory enough to investigate the influence of brine salting and smear-ripening on the iodine content in soft, semi-hard and hard cheeses under practical market conditions.

Increase of iodine content in brine-salted soft, semi-hard and hard cheeses by diffusion of iodide

Published online:

25 September 2019

Table 1. Composition of the experimental cheeses at the end of ripening.

Iodine contents in the edible part of soft, semi-hard and hard cheeses

The iodine contents obtained in the edible part of the experimental Camembert-, Tilsit- and Gruyère-type cheese are indicated in Table 1. The cheeses produced with non-iodised salt showed iodine contents in the range 30 to 57 µg kg⁻¹. The slightly higher iodine content of 57 µg kg⁻¹ in the Camembert cheese treated with non-iodised salt can be explained by the larger surface-to-volume ratio and mould growth, which increases moisture loss. In contrast, cheeses made with iodised salt had iodine content in the range 409 to 474 µg kg⁻¹. The obtained results show that brine salting and smearing of cheeses with salt supplemented with potassium iodide (KI) increased the average iodine content in the edible part of the three cheese types studied by a factor of about 10. The iodine content of the experimental cheeses treated with non-iodised salt was comparable to the iodine content of the processed milk of 35 µg kg⁻¹, i.e. the starting material. This is in good agreement with the findings of a previous study that showed that the iodine content of cheese is similar to the iodine content of the processed milk (Schöne et al. 2003). The low iodine levels of the cow's milk used in our study can be explained by the fact, that cheese production took place in October when iodine concentration in milk is low (Walther et al. 2018). A similar experiment was conducted with Camembert produced with cow's- and goat's-milk (Zimmermann et al. 2005). The iodine content in the cow's milk Camembert treated with non-iodised salt was up to five times higher than in our study (262 µg kg⁻¹ vs 57 µg kg⁻¹); however, it should be noted that the iodine content of 175 µg l⁻¹ in the processed cow's milk was also five times higher.

Zonal analyses of iodide and chloride in brine-salted semi-hard and hard cheeses

Soft cheeses such as Camembert are rather small and are eaten with the rind. For this reason, zonal analyses of soft cheeses were not completed. However, in the case of larger-sized cheeses such as Tilsit and Gruyère, which have been matured by smearing, the ripening process leads to the formation of a dry and tough rind. Therefore, a layer approximately 0.5 cm thick is usually removed before consumption. Zonal analyses were carried out during ripening, to study the diffusion of chloride and iodide in the semi-hard and hard cheeses in more detail. The iodide and chloride concentrations in rind, border and core zones (Figure 1) observed at different ripening times are shown in Figures 2 and 3, respectively. The cheeses treated with non-iodised salt contained low quantities of iodide, 27–54 µg kg⁻¹ in all zones (Figures 2(a) and 3(a)). In contrast, the cheeses treated with iodised salt had significantly higher iodide contents in all zones (Figures 2(c) and 3(c)). The results clearly showed that iodide from brine treatment and smearing migrated into the bodies of the investigated semi-hard and hard cheeses. However, the iodide concentrations in both cheese types decreased from the rind to the core zone. Three factors can explain the observed iodide concentration gradient: (1) the initial brine treatment using iodised salt, (2) the cure of the cheese surface with diluted brine, and (3) the progressive dehydration of the cheese rind by the release of moisture to ambient air during ripening.

Increase of iodine content in brine-salted soft, semi-hard and hard cheeses by diffusion of iodide

Published online:

25 September 2019

Figure 2. Iodide and chloride concentrations in the rind, border and core zones of Tilsit-type semi-hard cheeses after ripening times of 45 (white) and 90 (grey) days. Half of the four cheeses were treated with non-iodised salt (a, b), the other half with iodised salt (c, d). The figures e and f show the concentrations of iodide and chloride in the aqueous phase of the cheeses treated with iodised salt. The numbers in the graph represent mean values of duplicate samples.

Figure 2. Iodide and chloride concentrations in the rind, border and core zones of Tilsit-type semi-hard cheeses after ripening times of 45 (white) and 90 (grey) days. Half of the four cheeses were treated with non-iodised salt (a, b), the other half with iodised salt (c, d). The figures e and f show the concentrations of iodide and chloride in the aqueous phase of the cheeses treated with iodised salt. The numbers in the graph represent mean values of duplicate samples.

Increase of iodine content in brine-salted soft, semi-hard and hard cheeses by diffusion of iodide

Published online:

25 September 2019

Figure 3. Iodide and chloride concentrations in the rind, border and core zones of Gruyère-type hard cheeses after ripening times of 45 (white), 90 (grey) and 180 (black) days. Half of the four cheeses were treated with non-iodised salt (a, b), the other half with iodised salt (c, d). The figures e and f show the concentrations of iodide and chloride in the aqueous phase of the cheeses treated with iodised salt. The numbers in the graph represent mean values of duplicate samples.

Figure 3. Iodide and chloride concentrations in the rind, border and core zones of Gruyère-type hard cheeses after ripening times of 45 (white), 90 (grey) and 180 (black) days. Half of the four cheeses were treated with non-iodised salt (a, b), the other half with iodised salt (c, d). The figures e and f show the concentrations of iodide and chloride in the aqueous phase of the cheeses treated with iodised salt. The numbers in the graph represent mean values of duplicate samples.

For comparison, the chloride concentrations in semi-hard and hard cheese are also presented (Figures 2(b,d) and 3(b,d)). As expected, only small differences were observed in the chloride content between the cheeses treated with iodised and non-iodised salt. In contrast to iodide, the concentration of chloride increased from the rind towards the core zone. Chloride is mainly associated with the aqueous phase of cheese. Therefore, the lower concentration of chloride in the rind can be attributed to the lower water content in the rind. As a result of the repeated treatments of the cheese surface with diluted brine and the progressive dehydration of the cheese rind by releasing moisture into the ambient air, the solutes in the rind zone are continuously concentrated in the aqueous phase, which promotes their diffusion into the border and core zones.

All concentrations were related to the aqueous phase (Figures 2(e,f) and 3(e,f)), to verify differences in the diffusion behaviour between iodide and chloride. In the Tilsit-type semi-hard cheese an equilibrium of chloride through salt diffusion from the border zone to the core was obtained after 45 days (Figure 2(f)), whereas in the Gruyère-type hard cheese equilibrium of chloride was found after about 90 days (Figure 3(f)), which is in good agreement with previous findings for Gruyère cheese (Goy et al. 2012). Similarly, salt equilibrium was only reached after 83 days in hard Romano-type cheese (Geurts 1974). In contrast, even at the end of the ripening period, neither in semi-hard nor hard cheese was an iodide equilibrium achieved. As shown by the corresponding bar charts (c-f) in Figures 2 and 3, the migration of iodide was distinctly slower compared to chloride. One reason to explain the different migration behaviour is the larger size of the iodide ion compared with chloride that causes a greater obstruction to diffusion. The aqueous ionic radii for iodide and chloride are respectively 0.22 nm and 0.18 nm (Marcus 1983). The different migration behaviour is likely to be more pronounced in the rind and the border zones as these zones have a more compact structure due to their lower water content. Further to molecular iodide diffusion, interactions of iodide with the solid phase, solvent effects and the polarizability might play a role in the migration through the cheese matrix; however, such influences are difficult to quantify. Overall, only about 15% of the total iodide absorbed by the cheese remained in the rind, hence brine salting with iodised salt is an appropriate treatment for enriching iodide in the edible part of semi-hard and hard cheese. Hoffmann and co-authors reported similar findings in a study with Edam cheese treated with iodised salt containing potassium iodate (KIO₃). The iodine concentrations in the Edam cheese were at about the same level after 48 days of ripening as in our Tilsit cheeses after 45 days (Hoffmann et al. 1997).

Diffusion of iodide and chloride in cheese

The MFFB of cheese is an important criterion that affects not only firmness but also microbial growth, the intensity and rate of biochemical processes (e.g. proteolysis) and the velocity of physical processes such as the diffusion of water-soluble compounds (e.g. minerals, carboxylic or lactic acid) in the cheese body. The higher the MFFB, the faster concentration gradients of solutes are equilibrated in a cheese. However, due to various matrix effects, it is impossible to estimate the molecular diffusion of iodide and chloride in the aqueous phase of the cheese. Nevertheless, the movement of these solutes can be described as a diffusion process by using apparent diffusion coefficients (D_app). D_app is a convenient way to describe mass transfer processes through microporous matrices. The use of D_app implies that the mass transfer consists mainly of molecular diffusion, although other transport mechanisms may occur. As the moisture content of the cheese matrix is essential for diffusion, the experimental results of the rind (Figure 1) were omitted from the calculation of D_app due to the progressive drying of this zone during ripening. The D_app for iodide (I⁻) and chloride (Cl⁻) obtained from Equation (2) are listed in Table 2. Overall, the D_app for chloride calculated from the data of the present study were in a similar range to that previously reported (Guinee 2004) for sodium chloride in cheese moisture (1.2 × 10⁻¹⁰ to 5.2 × 10⁻¹⁰ m²s⁻¹). Figure 4 depicts the theoretical profiles of iodide and chloride quantities entering semi-hard and hard cheeses during ripening. The good agreement between the calculated profiles and the experimental data indicate that Fick's law (Equation 1) is suitable for modelling the diffusion of iodide and chloride within the aqueous phase of the semi-hard and hard cheeses.

Increase of iodine content in brine-salted soft, semi-hard and hard cheeses by diffusion of iodide

Published online:

25 September 2019

Table 2. Apparent diffusion coefficients (D_app) for iodide and chloride in semi-hard and hard cheese at ripening temperature (14–15°C), derived from the edible parts (without rind).

Increase of iodine content in brine-salted soft, semi-hard and hard cheeses by diffusion of iodide

Published online:

25 September 2019

Figure 4. Relative iodide and chloride uptake in function of diffusion time (M_t/M_∞) in semi-hard and hard cheese. The black and white dots represent experimental uptakes of iodide and chloride, respectively. The continuous and dashed line represent the calculated profiles for iodide and chloride, respectively, as calculated according to Equation (2).

Figure 4. Relative iodide and chloride uptake in function of diffusion time (M_t/M_∞) in semi-hard and hard cheese. The black and white dots represent experimental uptakes of iodide and chloride, respectively. The continuous and dashed line represent the calculated profiles for iodide and chloride, respectively, as calculated according to Equation (2).

The graphs in Figure 4 and the D_app listed in Table 2 show that chloride generally diffuses more rapidly into the interior of the cheeses than iodide, possibly due to the larger size of the iodide ion. Apart from the physicochemical characteristics (size, charge) of the solute ions, diffusion phenomena depend on cheese composition and structural properties. As shown in Table 2, there was a bigger difference between the two anions in semi-hard cheeses (ratio D_app 2.7) than in hard cheeses (ratio D_app 1.2). Significantly higher D_app values were obtained for chloride in semi-hard cheeses than in hard cheeses (ratio D_app 2.4), which can be explained mainly by the higher water content. In contrast, the observed iodide differences were much smaller (ratio D_app 1.1) between hard and semi-hard cheese, i.e. the matrix is physically less selective for the diffusion of this anion as its movement is retarded similarly by both cheese types. Thus, the occurrence of iodide-matrix interactions in the model cheeses is suggested. Due to the above-mentioned matrix effects, the determined D_app in cheese are considerably smaller than the molecular diffusion coefficients in aqueous solutions (298.15 K) of 2.03 × 10⁻⁹ m²s⁻¹ and 2.05 × 10⁻⁹ m²s⁻¹ for chloride and iodide, respectively (Buffle et al. 2007), which illustrates the extent to which diffusion is impeded. The study scope was limited to in-house produced semi-hard and hard cheeses; hence, the experimentally derived diffusion coefficients do not apply to products with different matrix compositions.

Contribution of cheese to salt and iodine intake of the Swiss population

Cheese is a valuable foodstuff consumed by many population groups. Particularly noteworthy is the high calcium content and the high abundance of essential amino acids. Except for some fresh cheeses such as Mozzarella, salt is used in the production of cheese. The salt content of ripened cheeses usually varies between 0.5 and 2.5% depending on the type and variety of cheese.

The experimental data of this study revealed considerable differences in the diffusion behaviour of iodide and chloride ions in smear-ripened cheeses. However, despite the slower diffusion of iodine, an average increase of 402 ± 30 µg kg⁻¹ (mean from Table 1) was achieved in the edible part of the cheese treated with iodised salt (Table 1). From a nutritional point of view, the inhomogeneous distribution of iodine in the edible part does not play a role in iodine supply, since the entire edible part is consumed.

In a Swiss long-term study, the contribution of cheese to the daily salt intake was estimated as 10% for women and 11% for men (Beer-Borst et al. 2009). According to this study, cheese is the second most important source of salt after bread. The mean total salt intake was 8.1 g per day in women and 10.6 g per day in men, which did not vary significantly over the years (Glatz et al. 2017) . Based on the results of these studies, a daily salt intake of 0.8 g (women) and 1.2 g (men) could be attributed to cheese. In Switzerland, iodised salt contains an iodine concentration of 25 mg/kg, which is added to the salt in the form of potassium iodide (KI). Assuming that the total uptake of iodine into cheese is proportional to salt and all cheese consumed is made with iodised salt, the iodine intake in adults could be increased by 20.3 µg d⁻¹ (women) and 29.2 µg d⁻¹ (men) corresponding to 14% and 19%, respectively, of the recommended daily iodine intake (RDI) of 150 mg/day for adults (WHO 2007). These theoretical considerations show that the use of iodised salt in the manufacture of cheese could be an important contribution to improve the iodine intake of populations in low iodine areas.

A recent study, the first National Nutrition Survey (menuCH) conducted from January 2014 till February 2015 provides representative information on the eating behaviour of Swiss adults aged 18–75 years from three linguistic regions of Switzerland (German, French and Italian) (Bochud et al. 2017; Chatelan et al. 2017). This survey collected data about the consumption of milk and dairy products (Benzi-Schmid and Haldimann 2018). According to these evaluations, adults aged 18 to 75 years consume on average 40 g cheese per day, with men having an average daily intake of 46 g and women 34 g. In both genders, the highest consumption was found in the 50 to 64 years age group. As the evaluation of the menuCH data did not distinguish between hard- and semi-hard cheese, we estimated the proportion of these two cheese types by the data provided by Agristat from 2016 (TSM Treuhand GmbH 2017). According to these data, the distribution of total daily consumption by cheese type is as follows: 11 g extra hard and hard cheese, 16 g semi-hard cheese and 13 g soft cheese.

Based on the iodine contents measured in the edible part of the experimental cheeses and the daily consumption of adults of soft (13 g), semi-hard (16 g) and hard (13 g) cheese, an iodine intake of 1.8 µg d⁻¹ (1.2% of the RDI) can be derived for cheeses with non-iodised salt. Accordingly, a value of 17.9 µg d⁻¹ (11.9% of the RDI) could be achieved for cheeses made with iodised salt. The use of iodised salt in cheese production could thus increase the daily intake of iodine for men to 20.5 µg (13.7% of the RDI) and for women to 15.2 µg (10.1% of the RDI). The age class with the highest cheese consumption (50–64 years old) would achieve a mean daily iodine intake of 20.2 µg (men 23.8 µg, women 16.3 µg) representing 15.9% and 10.9% of the recommended daily intake for men and women, respectively. In summary, these assessments based on different population-level dietary data suggest that a considerable improvement of the iodine supply to the Swiss population of at least 10% of the RDI would be possible through the consistent use of iodised salt in cheese production.

Conclusions

The data obtained in the present study from brine-salted semi-hard and hard cheeses show that iodide diffuses more slowly than chloride in the cheese matrix. As a result, the rind and border zone of the investigated experimental cheeses contained markedly higher iodine concentrations than the centre. Despite the uneven distribution within the cheese bodies, brine-salting of soft, semi-hard and hard cheeses using salt supplemented with potassium iodide is an effective measure to increase the iodine content in the edible part of various cheese types. However, depending on cheese manufacturing practice, matrix variability within the same cheese variety may influence the diffusion behaviour of iodine.

A high penetration rate of iodised salt in processed foods is important to ensure a properly balanced iodine supply for different population groups. Based on the results of the current work, the use of iodised salt in cheese production could make an important contribution to the iodine supply for the Swiss population, especially for high risk groups, such as pregnant women. Cheese could be an important nutritional source of iodine in Switzerland; however, the current use of non-iodised salt in its production reduces its contribution to iodine intakes in the population. It therefore seems desirable to use iodised salt in brine salting.

In Cheese Making Why Can You Never Use Iodized Salt

Source: https://www.tandfonline.com/doi/full/10.1080/19440049.2019.1668571

Share :