The present invention relates to a process and a material for the inhibition of gas formation in fish ensilage.

For a long time, there has been a great problem within the fish industry in connection with production and transport of fish ensilage that large amounts of unidentified gas are formed in the ensilage.

Fish ensilage is a product which is made by addition of acids such as formic acid and sulphuric acid to the whole fish or parts thereof. The product may be purified or fractionated to increase the quality, and the ensilage can be dewatered to different extents.

The problem with gas formation is experienced by many ensilage suppliers. The problem has been most prominent and occurred most often during the summer months, and it is believed up until now that the gas formation was associated with the amount of fish bones which were admixed with the ensilage.

Under acidid conditions and with a large content of fish bones, one could risk that the carbonate in the fishbone would dissolve with liberation of CO2 and consumption of protons with subsequent increase in pH, according to the following reaction: CaCO3 + 2H+-> Ca 2+ + CO2 + H2O The inventors of the present invention have, however, studied the problem with gas formation in fish ensilage in more detail and our preliminary conclusion is that other factors contribute significantly to the gas formation. As will be apparent from the experiment given below, we have

concluded that micro-organisms are the source of the gas formation and thus, the present invention provides methods and materials to inhibit microbial growth in the fish ensilage.

Thus, the present invention comprises a method to prevent gas formation in fish ensilage characterised in that an effective amount of an agent which inhibits microbial growth is added to the fish ensilage.

In a currently preferred embodiment the agent is sodium benzoate.

The present invention comprises also a method for preventing gas formation in fish ensilage, characterised in that manufacture and storage of fish ensilage take place at a reduced temperature, preferable at about 5 °C.

Furthermore, the present invention comprises a fish ensilage fluid, characterised in that in addition to the usual components it comprises an agent which inhibits microbial growth.

A number of experiments were carried out to identify and solve the problem. As no clear understanding existed of what the cause of the gas formation was, the strategy was at first to characterise the ensilage chemically, as well as microbiologically in addition to determining which gas was produced. In a commission from Hordafór AS, Norconserv, represented by Torstein Skara, carried out the experiments.

Example 1 Gas formation-ensilage.

Samples of a bone-rich ensilage were taken at Sotra Fiskeindustri on June 10t, 1998. Three samples were wrapped in plastic bags and sealed under vacuum.

One sample (no. 1) was stored at room temperature, another (no. 2) was incubated at 20 °C and a third at 5 °C.

In the samples nos. 1 and 2, the gas formation was noticeable (many times the volume of the ensilage sample) after 2-3 days. Preliminary gas samples taken from sample no. 1 showed that the gas did not contain CO2. This was, however, invalidated by analysis of gas from sample no. 2, on a GC-MS and an external gas analysis instrument-which reported >50% CO2. Therefore, we have reason to believe that the gas, in the main, consists of CO2.

Example 2 Composition-Analysis results.

According to the result from example 1, we could not give any precise explanation of the cause of the gas formation in the ensilage. Therefore, an analysis of the ensilage itself was carried out, which gave the following results in sample 2: pH 3.5 Colonies 130 cfu/g Mould <10 cfu/g Lactic acid bacteria <10 cfu/g Yeast <10 cfu/g The results from the analysis show that the pH in the ensilage is so low that we expect it to have preserving properties in itself. On the other hand, because of the large amounts of nutrients which exist in ensilage, one could expect microbial growth and activity. Generally, one would expect that lactic acid bacteria and/or yeast could grow at low pH values. The results show that this is not the case for the analysed sample.

However, one should not exclude that specially adapted bacteria would be able to adapt to this acidic environment.

Example 3 Gas production in ensilage. Change in storage temperature.

Sample no. 3 was stored at 5 °C as described in example 1. No gas production was detected after storage for 14 days at this temperature.

After 14 days, the temperature was increased to 20 °C.

After a couple of hours, significant development of gas was apparent in the bag, as around three times as much gas (by volume) as the volume of the ensilage was formed. It was assumed that this gas formation was the chemical process described above as the bag with the test material was sealed such that no new micro-organisms could enter. The wrapping of the ensilage in closed packing could contribute to stabilising the equilibrium: 2H + C03 H2C°3 H20 + C°2 From this experiment, however, we cannot exclude that growth of micro-organisms is the cause, as the reason for low, or absence of, gas formation at 5°C can be due to the prevailing micro-organisms having low activity and growth rate at this temperature, even if this will be somewhat surprising.

Example 4 Addition of preserving agent-sodium benzoate.

To exclude microbial causal connections, an experiment was carried out with a preserving agent added to the ensilage. If the preserving agent had an effect on the gas production, it would be a clear indication that the problems were caused by microbial growth.

Again, a sample of ensilage was collected from Sotra Fiskeindustri on September 22nd, 1998. The sample was divided into 5 parts of 200 ml and wrapped in plastic bags under soft vacuum. Into two of the bags, 5 g benzoic acid was added. Two bags, one with and one without benzoic acid were then incubated at 20 °C. After a few days, gas formation could be observed in the bag with no benzoic acid added. There was approximately 3 times as much gas formed (by volume) as the volume of the ensilage. There was, however, no gas formation in the bags with sodium benzoate added. It seemed therefore, that the gas production was associated with microbial growth.

Example 5 Identification of gas producing organism.

To determine if the gas formation could really be explained by microbial growth, a sample from one of the bags in which gas had formed was plated out. Because of the low pH in the ensilage, samples from the ensilage were also plated out on media having pH 3.5 and 5.5. Examination of the plates under a microscope gave no clear indications of causal connections, and cultivation of plates gave no growth of either lactic acid bacteria or yeasts.

Cultivation under anaerobic conditions gave clear indications of gram positive rods, and initially we assumed that this could be Clostridia or Leuconostocs. Isolated colonies were picked out and identified, and we have in the first instance concluded with the gas formation being due to bacteria species from Clostridium, and probably Clostridium botulinum.

Example 6 Effect of various concentrations of sodium benzoate on the formation of gas in ensilage.

Samples of ensilage were collected from Sotra Fiskeindustri on April 26th, 1999. The samples were packed

in portions of 200 ml (95% vacuum). Two units of each variant were packed, as shown in Table 1.

Table 1: Survey over added amounts of conserving agents and storage temperature AdditiveAmountStoragetemp. Sodiumbenzoate 2g (lOg/kg), lg 20 C (5g/kg),0.4g(2g/kg) 0,2g(lg/kg) Potassiumsorbate2g(lOg/kg),0.2g 20 li c (lg/kg) Control20 C 0 20 C Control 4 C 0 4 C After 2 days, gas formation was apparent in the packs with no added preserving agent and stored at 20 °C. There were also signs of gas formation (bubbles) in the packs to which were added 1 g/kg and 2 g/kg respectively, while the packs with more sodium benzoate were without signs of gas formation.

After 5 days about 1 litre of gas was formed in both the bags with ensilage with no added preserving agent, while there were few or no signs of gas formation in any of the other bags. Therefore, we cannot exclude that levels of sodium benzoate lower than 1 g per kg ensilage could also be sufficient to prevent gas formation at 20 °C.

The conclusion is that both sodium benzoate and potassium sorbate inhibit gas formation in ensilage.

Example 7 The above experiments were carried out with fish ensilage with formic acid being used as an ensilage agent.

We have also carried out tests where sulphuric acid/acetic acid was used as ensilage agent, and gas formation was a considerable problem in this ensilage too.

2.0 % sulphuric acid and 0.5 % acetic acid were added to fish (denoted industrial fish). The pH in the ensilage was 3.0.

Sodium benzoate was added to this ensilage and there was no noticeable gas formation in this test, as opposed to in the control test.

From the experiments above, we have surprisingly shown that the gas formation is a consequence of microbial activity, and is probably caused by the identified species of Clostridium bacterium. From this, we would expect that

all preserving agents, i. e. agents which prolong the lasting qualities of foodstuffs by inhibiting deterioration caused by micro-organisms, would inhibit gas formation in ensilage.

Furthermore, it seems that gas formation is temperature dependent, so that the problem is more prevalent during the summer months. Obviously, a possibility to reduce gas formation will be to keep the temperature at, for example, +5 °C. Practically, this is not unproblematic and it may seem simpler and less resource demanding to use a preserving agent.

Sodium benzoate is on the « positive list » for fish and other aquatic feed materials from 1991. Benzoic acid is poorly soluble, therefore salts are often used, e. g. benzoate. Benzoic acid has better bacteria-inhibiting properties than sorbic acid because it contributes to preventing the bacteria from using the water phase in the various materials present. It also inhibits yeasts, but is less effective when it comes to inhibiting moulds. Benzoic acid is most effective in the pH range 3-4, but can also be used in weakly acidic materials up to pH 6. In connection with foodstuff, it is approved for use with fish, herring and shellfish products, jams and marmalades, vegetable products, soft drinks and juice.

Thus, one can imagine many different solutions: 1. After the ensilage has started to « boil » i. e. form gas, addition of an agent which kills micro-organisms, such as a bacteria-killing agent, will stop the growth of the gas forming bacteria, and thus the gas production. In this way one can imagine a site with a stand-by situation in which, for example, sodium benzoate is added as the « boiling » starts, i. e. when it is necessary.

2. One can ensure against the phenomenon by adding sodium benzoate directly to the ensilage fluid, so that a small amount of bacteria-killing agent is present in the stock all the time during the ensilage process. For example, such that the concentration of sodium benzoate in final ensilage is ca. 5 g/kg.

3. Furthermore, we have in the experiments, which are explained above, shown that the gas formation is temperature dependent and the production and transport of ensilage at reduced temperature will therefore inhibit the gas formation.

4. A combination of (a) reduced temperature and (b) addition of an agent which inhibits growth of micro- organisms can be used.

We have started testing a number of different preserving agents to find out which is best suited, and the concentrations and temperatures which are most expedient to use. The tests are carried out both in the laboratory and on a large scale. We want to mention, however, that the solution to the problem was in recognising that micro- organisms, and probably bacteria, are the real cause of the gas formation. When the causal connection is proven, it is relatively simple for those skilled in the art to find the preservation agents that ought to be used, including the concentration ranges, incubation conditions, etc.

Thus the invention will not be limited to the examples given above, but the scope of the invention will be defined as it is described in the following claims.