The present invention refers to a core-shell structure, a method for making a core-shell structure and the use of a plurality of core-shell structures according to the independent claims.

Piglets are highly susceptible to enteric disorders immediately after weaning. They then have an immature digestive system, the capacity of which for nutrient digestion is impaired by the transition from sow milk to a solid diet. Diarrhea is a common post-weaning disease that is characterized by reduced digestive capacity leading to poor growth performance. Because of the im ^¬ mature microflora, proliferation of enteropathogenic bacteria has a high incidence that further predisposes piglets to enteric infections.

In order to mitigate these effects, weaned piglets were fed in the past with diets routinely supplemented with subtherapeutic levels of antibiotics as growth promoters (AGPs) . However, AGPs are suspect of increasing the antibiotics resistance of patho ^¬ genic bacteria, which is a potential risk to human health. AGPs have therefore been banned from swine production in many countries and are under pressure to be banned globally. These developments have sparked considerable interest in identi ^¬ fying alternative nutritional strategies in pig feeding. Various different approaches have been developed relying on novel feed additives or alternative feeding strategies. An overview of the progress in this field has been provided in two review articles (H. H. Stein, Feeding the Pigs' Immune System and Alternatives to Antibiotics, Proceedings London Swine Conference - Today's Challenges - Tomorrow' s Opportunities 2007, 65-82; M. Nyachoti, J. M. Heo, Feed Additives and Feeding Strategies to Replace An ^¬ tibiotics, Advances in Pork Production 2013, 24, 123-127) .

A widely employed alternative to AGPs is the supplementation of the pig's diet with pharmacological levels (3000-4000 ppm) of zinc oxide (ZnO) . Zinc oxide shows similar effects on daily growth and feed conversion rates as antibiotics employed as growth promoters. Thus, zinc oxide fed at pharmacological levels can be a cost-effective strategy for controlling post weaning diarrhea. The approach does not affect the resistance of bacte ^¬ ria. However, with such high levels of zinc oxide, concerns about adverse environmental effects (accumulation of heavy metal salts in the environment) have arisen. This has led to a re ^¬ strictive use of this method. According to regulation No.

1831/2003 by the European Commission, 150 ppm of zinc is allowed in compound feedstuff in the European Union. Switzerland knows the same maximum value for young farmed animals (cf . Liste der zugelassenen Futtermittelzusatzstoffe (Zusatzstoffliste 2)). Orffa Additives B.V. (NL) distributes micro-encapsulated zinc oxide for the prevention of post-weaning diarrhea in piglets. It is surmised that, when passing the piglet's stomach, zinc oxide is partially converted into zinc chloride which does not show the desired properties. By micro-encapsulation with a vegetable fatty acid matrix, zinc oxide is prevented from these reactions. When entering the intestines, the coating is digested by lipase and bile salts, and zinc oxide is released. Due to its increased efficacy, micro-encapsulated zinc oxide can be used at levels a ten-fold lower than common inorganic zinc oxide, maintaining similar results on diarrhea prevention and production performance. However this is still not sufficient to meet the new reg ^¬ ulations . It is therefore a problem underlying the present invention to overcome the above-mentioned drawbacks in the prior art. In par ^¬ ticular, it is a problem underlying the present invention to provide a feed additive or feedstuff, which shows similar or even better growth and feed conversion rates than those achieved with subtherapeutic antibiotics or pharmacological levels of zinc oxide in the feeding of weaned piglets. More particularly, the heavy metal concentrations that are to be achieved in a com ^¬ pound feedstuff are less than 100 ppm. This is in line with cur- rent regulatory allowances.

This problem is solved by a core-shell structure, a method for making a core-shell structure and the use of a plurality of core-shell structures in a feed additive or feed stuff compris- ing the features in the independent claims.

The core-shell structure comprises a core of titanium oxide and a shell essentially comprising, preferably consisting of, an ox ^¬ ide of an element selected from a group consisting of zinc, cop- per, vanadium, chromium, manganese, cobalt, molybdenum, tungsten, silver and mixtures thereof.

By providing these heavy metal oxides in the form of a core- shell structure, their concentration in a feed additive or feed stuff can be significantly reduced without diminishing their ac ^¬ tivity. Similar growth and feed conversion rates as those achieved with high dosages of zinc oxide reported in the litera ^¬ ture can be achieved at substantially lower heavy metal load ^¬ ings .

The core-shell structure can have a specific surface of at least 20 m ² /g, preferably at least 60 m ² /g, more preferably at least 120 m ² /g. In certain embodiments, the core-shell structure can have a specific surface between 20 m /g and 200 m /g, in particu ^¬ lar between 40 m ² /g and 160 m ² /g or between 60 m ² /g and 120 m ² /g. Such high specific surfaces can be achieved with porous shells. This allows for a further increase of the active surface of the heavy metal oxides and therefore for even better growth and feed conversion rates at lower dosages.

In the context of the present invention, specific surfaces are determined with the BET-method, specifically according to the norm DIN-ISO 9277.

In the context of the present invention, the core shell- structure further comprises a hydrophobic coating, preferably on top of the shell, in particular of at least one lipid. It has been found that core-shell structures of the above-mentioned kind exhibit a significantly better growth and feed conversion rate, if they bear a hydrophobic coating. Therefore the heavy metal loadings can be further reduced and the results achieved are superior to those attained with subtherapeutic antibiotics or pharmacological levels of zinc oxide. The heavy metal concen ^¬ trations can be reduced to less than 100 ppm in a compound feed, which is well in-line with current regulatory allowances.

However, there is also disclosed a core-shell structure as de- scribed hereinabove, which comprises no hydrophobic coating.

The at least one lipid can be a fatty acid, in particular an un ^¬ saturated or a saturated fatty acid. Fatty acids show high bio- compatibility and, hence, a better growth and feed conversion rates. The fact that either unsaturated or saturated fatty acids can be used allows for adapting the kind of fatty acid employed to other constituents of a compound feedstuff. The fatty acids can be selected from a group consisting of myristoleic acid, palmitoleic acid, sapienic acid, oleic acid, elaidic acid, vac- cenic acid, linoleic acid, linoelaidic acid, -linolenic acid, arachidonic acid, eicosapentaenoic acid, erucic acid, docosahex- aenoic acid, caprylic acid, capric acid, lauric acid, myristic acid, palmitic acid, stearic acid, arachidic acid, behenic acid, lignoceric acid and erotic acid.

In a plurality of core-shell structures of the above-mentioned kind, the diameter of 90% of the cores can be in a range between 0.01 μιη and 100 μιτι, preferably between 0.1 μιη and 10 μιτι, more preferably between 0.15 μιη and 0.75 μη. With this particle size distribution, optimal results are achieved.

Disclosed is also a method for making a core-shell structure. The method comprises the steps of:

- Providing a core of titanium oxide;

- Treating the core with an ammonium salt and a metal salt, in particular a salt of a metal selected from a group consist- ing of zinc, copper, vanadium, chromium, manganese, cobalt, molybdenum, tungsten, silver and mixtures thereof, to obtain an intermediate;

- Calcining the intermediate to obtain a core-shell structure;

- Optionally: Coating the core-shell structure obtained with a hydrophobic coating, preferably on top of the shell, in particular of at least one lipid, to obtain a coated core- shell structure.

The described method allows for efficient production of various core-shell structures. Furthermore, the method is highly versa ^¬ tile, as a large range of different structures can be produced by employing different combinations of starting materials. The present invention further refers to a method for making a core-shell structure as described above. The method comprises the steps of: - Providing a core of titanium oxide;

- Treating the core with an ammonium salt selected from a

group consisting of ammonium carbamate and ammonium carbonate and with a metal salt, in particular a salt of a metal selected from a group consisting of zinc, copper, va- nadium, chromium, manganese, cobalt, molybdenum, tungsten, silver and mixtures thereof, to obtain an intermediate;

- Calcining the intermediate to obtain a core-shell structure;

- Coating the core-shell structure obtained with a hydrophobic coating, preferably on top of the shell, in particular of at least one lipid, to obtain a coated core-shell struc ^¬ ture .

The above-mentioned ammonium salts lead to intermediates that are particularly susceptible to calcination. The metal salt used can be a halide, preferably a chloride. These are abundantly available at low prices.

In such a method, a plurality of cores of titanium oxide can be provided and the diameter of 90% of the cores can be in a range between 0.01 μιη and 100 μιτι, preferably between 0.1 μιη and 10 μιτι, more preferably between 0.15 μιη and 0.75 μιη.

In the above-described methods, the step of treating the core with an ammonium salt and a metal salt can be conducted in pres- ence of a template. Such a template leads to an increased sur ^¬ face area of the shell. Preferably, the template can be a fat suspension . The present invention further refers to a food additive, a feed additive or a feed stuff comprising a plurality of core-shell structures as described above.

Such a food additive, feed additive or feedstuff can comprise core-shell structures with copper and zinc, either separate or in combination, wherein the ratio between copper and zinc is in a range between 1:100 and 1:1, preferably between 1:50 and 1:2, most preferably between 1:20 and 1:5. With such a combination, good results can be achieved with respect to growth and feed conversion rates at particularly low heavy metal loadings.

The present invention further refers to the non-therapeutic use of a plurality of core-shell structures of the above-described kind as a food additive, feed additive or in a feedstuff. Howev er, it has to be noted that the use of such core-shell struc ^¬ tures is by no means restricted to such applications. By way of example, core-shell structures of the above mentioned kind may also be employed in cosmetics.

Furthermore, a plurality of core-shell structures according to the present invention can also be for use as a medicament, in particular for use in the treatment of diarrhea or food poison ^¬ ing .

Further advantages and features of the present invention become apparent from the following discussion of several embodiments and from the figures. It is shown in

Figure 1 : Schematic representation of a core-shell

structure according to the present invention;

Figures 2 to 11: Experimental results of in vivo studies con ^¬ ducted with several core-shell structures. A variety of core-shell structures were prepared and tested in vivo. For illustration, the preparation of a structure with a shell of ZnO and a coating of fatty acids having an overall zinc content of 10% is discussed in the following:

T1O ₂ (5 kg) was added to water (20 1) . The resulting mixture was stirred vigorously until it became homogenous. Ammonium carba ^¬ mate (630 g) was added and stirring was repeated until the mix ^¬ ture was homogenous again. A solution of ZnCl ₂ (4325 g) in water was added and vigorous stirring was continued, during which gas- evolution and the formation of a foam-like gel was observed. Af ^¬ ter gas-evolution had ceased, the gel was allowed to stand for several hours. The mixture was filtered and the resulting resi ^¬ due was washed with water (three times) . The solid obtained was dried at 120° C until its weight was constant and then subjected to calcination at 400° C. The calcinated material was sieved and coated with a mixture of fatty acids (C ₁₂ to C ₂ 0 saturated and un ^¬ saturated fatty acids) by spraying the material with the mix ^¬ ture .

Upon addition of ZnCl ₂ to the mixture, the following reaction oc ^¬ curs :

5ZnCl ₂ + 5NH ₂ C0 ₂ NH ₄ + 8H ₂ 0 → Zn ₅ (OH) 6 (C0 ₃ ) 2 (s) + 3C0 ₂ (g) + 10NH ₄ C1

Due to the CO ₂ gas generated, the Zn ₅ (OH) ₆ (CO3) ₂ forms as a foam ^¬ like gel that settles on the titanium oxide. After filtration and drying, the coated T1O ₂ is subjected to calcination in order to obtain a highly porous coating of zinc oxide:

Zn ₅ (OH) ₆ (C0 ₃ )2 (s) → 5ZnO (s) + 2C0 ₂ (g) + 3H ₂ 0 The core-shell structures obtained can optionally be coated with a hydrophobic material, in particular with a lipid, such as a fatty acid. By employing this method, various materials accord ^¬ ing to the present invention were prepared with different heavy metals, contents thereof and coatings.

In the above procedure, a template was optionally added in order to increase to surface area of the zinc oxide. The template was prepared by mixing sunflower oil (100 g) , Tween 20 (10 g) and octanol (10 ml) in a blender. The resulting mixture was added to the T1O ₂ slurry together with the ammonium carbamate. The

Zn ₅ (OH) 6 (CO3) 2 was thus formed in presence of the template. During calcination, the template was degraded to afford the zinc oxide in highly porous form.

Figure 1 provides a schematic representation of the core-shell structure 1 prepared in the above-mentioned example. The core 2 of T1O ₂ is coated with a shell 3 of porous ZnO. Furthermore, the whole particle is coated with a layer 4 of fatty acids. The car- boxylic acid residues of the fatty acids are presumed to adhere on the surface of ZnO, rendering the exterior of the core-shell structure hydrophobic. It is reasonable to assume that this hy ^¬ drophobic coating protects the structure, in particular the lay ^¬ er of ZnO, against gastric acid when passing the pig' s gastric- intestinal tract. The coating is then presumably degraded in the intestine, setting free the uncoated particle.

Table 1 provides an overview of the materials subjected to in vivo testing: Table 1: Overview of materials subjected to in vivo testing

In these experiments, the core-shell structures were used as a feed additive for weaned piglets. Each run was conducted with two groups of about 50 animals, which were fed with two differ ^¬ ent feedstuff compositions. The results obtained were compared to those of a control group ("Null") , which obtained the same diet, but without additive. The weight of the animals was rec ^¬ orded at regular intervals. The results shown in the figures re- fleet the weight after approximately one month of feeding. The dosage of feed additive was always 500 g per ton feedstuff. This generally corresponds to an overall heavy metal content of 50 ppm. Figures 2 and 3 summarize the results of the first experiment, in which two uncoated core-shell structures were tested. The first structure (L-Zn) was based on a core of T1O ₂ with a coating of ZnO. The material had a zinc content of 10%. The second mate ^¬ rial tested (H-Zn) had the same structure, but had a zinc con- tent of 20%. The health condition of the animals was fair, which can be considered as normal for piglets after weaning. As can be seen from Figure 2, both feed additives led to a slightly diminished daily growth rate (the higher, the better) compared to the control group. However, the results in Figure 3 clearly show that an im ^¬ proved feed conversion rate (the lower, the better) was

achieved. Surprisingly, the material having the lower zinc content (L-Zn) showed a more pronounced effect than the material with the higher zinc content (H-Zn) .

Figures 4 and 5 show the results of Experiment 2. Here, the core-shell structures with the low zinc content of 10% (L-Zn) were coated with two different oils, namely paraffin oil (P) and sunflower oil (S) . The overall amount of zinc in the supplement- ed feedstuff was 50 ppm in both cases. The health condition of the piglets fed was comparably good. It becomes apparent from figure 4 that both materials tested still led to a lower daily growth rate than observed with the control group. However, the material coated with sunflower oil only showed a slightly nega- tive effect. On the other hand, as can be seen in Figure 5, a significantly improved feed conversion rate was observed with both materials.

Figures 6 and 7 summarize the results of Experiment 3. Here, the core-shell structures coated with sunflower oil (S) were com ^¬ pared to an analogous material with a fatty acid coating (FA) . The overall health condition of the animals was only fair. Pre ^¬ sumably, this is the reason as to why an improved daily growth rate compared to the control group was observed this time with both samples, which becomes apparent from Figure 6. Figure 7 re ^¬ veals that, in this case, the structures coated with sunflower oil led to a feed conversion rate comparable to that of the con ^¬ trol group. Surprisingly, a significant improvement of the feed conversion rate was observed for the core-shell structures coat ^¬ ed with fatty acids.

Figures 8 and 9 summarize the results of Experiment 4. Here, the core-shell structures of Experiment 3 coated with fatty acids were compared to a feed additive consisting of fatty acid-coated core-shell structures based on ZnO or CuO, respectively. The content of zinc and copper was 10% in both kinds of material. In the feedstuff formulation, 95% of zinc-based structures were used with 5% of copper-based ones. This again corresponds to an overall heavy metal content of 50 ppm in the feedstuff.

The health condition of the animals was fair. As apparent from Figure 8, an increase of the daily growth rate could be achieved with both materials. Figure 9 reveals that a slightly inferior feed conversion rate was observed with the material that was ex ^¬ clusively based on zinc oxide. In contrast, the combination of zinc and copper led to a significant improvement of the feed conversion rate. Indeed, the feed additive based on the oxides of copper and zinc led to results, which were clearly superior to those normally observed with pharmalogical levels (3000 to 4000 ppm) of conventional zinc oxide.

Figures 10 and 11 show the results of Experiment 5. The same ma- terials as in the previous one were tested here, but the overall health condition of the animals was poor this time. Figure 10 indicates that both feed additives led to an increase of the daily growth rate. As shown in Figure 11, no significant differ ^¬ ence in the feed conversion rate between the materials was ob- served this time. Apparently, the beneficial effect of copper on the feed conversion rate only has an influence, if the health condition of the animals is compromised. In conclusion, a feed additive is provided, which leads to simi ^¬ lar or even better growth and feed conversion rates than those achieved with subtherapeutic antibiotics or pharmacological lev ^¬ els of zinc oxide. Most notably, the heavy metal concentrations in compound feedstuffs are well below 100 ppm.