GB2324299A - Producing glass ceramic from ferrosilicate powder - Google Patents

Abstract

A method of producing glass ceramic from ferrosilicate powder comprises heating the powder to a glass-forming temperature in the presence of at least one additive, such as glass or sodium oxide or sodium carbonate, to reduce liquidus temperature. In a modification, ferrosilicate slag is incorporated in a synthetic plastics matrix.

Description

This invention relates to ferrosilicates and ferrosilicate products.

Ferrosilicate powder and slag are produced as by-products of processes such as copper refining where iron is used as a flux to remove silicon impurities. The so-called slag residue is principally dumped as waste, although some is coated in copper and used in agriculture and sane of the coarser slag used for shotblasting. However, even the material used for shotblasting is eventually dumped once it has become too fine for further use.

The present invention envisages using this waste material to constructive purpose either by direct inclusion into products or by inclusion subsequent to modification.

In particularly preferred embodiments of the invention the powder or slag is modified to provide or enhance magnetic properties prior to incorporation in other materials such as glasses or plastics.

The invention is now described by way of example with reference to the accompanying examples.

In the accompanying description, the invention is illustrated with reference to 'copper slag'. However ferrosilicate slag from other sources may also be utilized even if the composition varies.

Ferrosilicate slag produced in copper refining, or finely divided after shotblasting, essentially comprises a mixture of silicates and iron in its lower, ferrous, oxidation state.

In considering how ferrosilicate slag could be utilized, it was decided to attempt to convert the ferrous oxide into ferric oxide, a material of higher value which also exhibits ferromagnetic properties.

The various slag materials available have different compositions depending upon their source, that is their production source and also any subsequent use such as shotblasting. To be useful, a higher consistency is desirable, which can be achieved through mixing. It was decided to combine the mixing process with heating to see if a glass could be made using the silicates in the mixture and also convert at least some of the iron oxide. Suitable adaptation of the heating process would then be employed in order to modify the magnetic properties.

Initial melting of ferrosilicate on its own was not satisfactory, requiring high temperatures in excess of 14500C and not producing a pourable glass. The melt produced did not dissolve all the material, leaving a refractory scum, so that homogeneity was not obtained.

Containing the melt was also problematic as it was very corrosive.- To increase percolation and dissolution, glass cullet and sodium oxide (from sodium carbonate) were added. By selecting suitable quantities of cullet a pourable glass was obtained at lower temperatures, approximately in the range of 1350 to 14000 which is a workable range in glass manufacture. As sodium carbonate was added, the scum dissolved. Preferred weight percent additives are of the order of up to 10% for each of the glass cullet and sodium carbonate, but up to about 20% could be used. Higher percentages of additive are generally avoided if the objective is to utilize as much of the ferrosilicate slag as possible.

On cooling the melt produced with the glass cullet and sodium carbonate, the melt devitrifies into a glass ceramic comprising ferric oxide (magnetite) in a matrix of silica and other trace materials.

Within the context of this specification the expression 'glass ceramic' is used broadly to imply the presence of particles, normally the majority material, within a glass matrix, normally the minority material of the composite. In the present invention the volume percent of the magnetite or particle inclusions may be up to 80%, preferably in the range of 70 to 80%, which is not as high as the percentage of inclusions in same glass ceramics.

The melted phase of the material may be cast or handled and formed as any glass.

Specific examples of procedures are now given.

Starting Materials Two grades of ferrosilicate powder were used as starting materials. The first, called herein 'powder D', was a fine, black, free flowing powder having particle diameters less than Inn. The second, called 'powder S', was initially a very wet powder that appeared coarser than powder D. Powder S was dried to form a free flowing powder and had particle sizes ranging from very much less than Inn up to around 4mm.

The main component of both powders was black, angular materials having clean, shiny fracture surfaces implying an amorphous nature; it is believed both black powder materials were an iron silicate glass. Both powder contained smaller amounts of white, sandy grit, red brick dust and organic matter as impurities.

Some dried powder S was sieved through a 710cm sieve. The sieved material appeared visually similar to the unsieved powders.

This procedure yielded three starting powders, powder D, powder S sieved and powder S unsieved. X-ray diffrction of the powders showed them all to be mainly amorphous with a track of fayalite, FC2SiO4, and other minor phases one of which may be magnetite, Fe3O4. All samples had two unidentified peaks that were much enhanced in powder S sieved, while powder S unsieved showed a quartz peak and powder D traces of a-Fe. All three powders were magnetic.

Glass Melting Prior to glass melting and to avoid thermal shock when being introduced into the furnace, the crucibles containing the raw materials were sintered at 10000C overniffl . It was observed that the colour of the raw material changed-Yfrom black to brown. This was accompanied by an increase in weight. It was concluded that during the sintering Fe particles were becoming oxidised and that this was responsible for the weight gain and brownish hue.

The raw material with no additions provided difficult to melt, requiring temperatures in excess of 14000C. At this temperature corrosion of the alumina cricible was often observed which resulted in loss of the melt product. In order to reduce melting temperature, quantities of cullet (broken bottle glass) and Na2CO3 were added systematically to the starting powder. These two additions were thought to be suitable to reduce the liquidus temperature. In addition, a greater fluidity in the molten state would help to dissolve some of the scum, observed on the surface of the initial melts.

Adding of Na2 CO3 10 and 20 wt% Na2CO3 was added to the melt. This was successful at reducing the melting temperature, thus preventing corrosion of the crucible. In addition, the scum at the surface became incorporated into the melt. 20 wt% Na CO however was considered too much since the fluidity of 23 the melt was so high that pouring became difficult.

Addition of Cullet Up to 30 wt% cullet was added to the batches in order to reduce the melting temperature. This was also successful except that the amount of cullet necessary to have the same effect as 10 wt% Na2CO3 was higher (20 wt%).

Analysis of Cast Material It was observed fairly easily that, in the cast material, all but the most rapidly quenched lower surface had a faceted appearance that would normally be associated with a crystalline phase. X-ray diffraction of a typical boule revealed that this phase was magnetite, giving an almost perfect match to the X-ray data files. All samples contained this phase although it is quite clear upon examining the fracture surface of the boules that the material with 30 wt% cullet retained its amorphous structure more easily than other compositions.

All samples exhibited strongly ferramagnetic behaviour.

Addition of Fe Since it was concluded that the material was primarily an iron silicate, the iron oxide - SiO2 phase diagram was considered to be useful guide to melting behaviour. A eutectic exists in this phase diagram at approximately 70 mol% Fe304. Since the compositional analysis suggested that the maximum magnetite content with the raw material was 60 mol%, Fe was added to the batch prior to melting. This had the effect of decreasing the liquidus and improving the fluidity of the melt. However, the as-cast material only gave a weak ferromagnetic response. It was concluded that the boule was either amorphous or had crystallised as FeO rather than Fe3O4. This was thought to relate to the partial pressure of oxygen in the melt and could be corrected with more research.

Using a combination of the supplied raw materials and small additions to improve glass forming and melting behaviour, strongly ferromagnetic devitrified boules could be produced.

Ferrosilicate slag produced in copper refining, or finely divided after shotblasting, essentially comprises a mixture of silicates and iron in its lower, ferrous, oxidation state. Although it is known to incorporate a variety of waste materials of differing types into plastics, either for dumping purposes or for subsequent use in the construction industry, it is has not been considered possible to use slag for this purpose because of copper contamination which is detrimental to the properties of plastics. However it has surprisingly been found possible to incorporate raw slag into plastics materials including Nylon, polypropylene, styrenes and many others of both thermoplastic and thermosetting varieties, without impairing their properties.

Even more surprisingly, it has been found possible to incorporate very high percentages (i.e. over 50%) of slag into the plastics matrix and that the resulting materials exhibit unusually high strength compared with the same matrix incorporating similar quantities of other fillers.

For example, in Nylon the slag provides superior strength to the incorporation of glass beads and in polypropylene up to 94% ferrosilicates can be incorporated. The material is also comparatively light, providing a very high strength to weight ratio.

Ferrous oxide can potentially be converted to ferric oxide, which has soft magnetic properties. By subjecting the ferrosilicate slag to a heating process it was^-found possible to make this conversion to the iron oxide within the slag, with the result that the resulting product exhibited magnetic properties. The conversion potentially takes place at around 7000C, but it was found necessary to utilise higher temperatures of the order of 13000C to melt the silicates to enable the process to take place in bulk slag material.

The heating process has drawbacks in that the molten slag is extremely corrosive and the end product after resolidifying is a solid mass that requires grinding prior to use as a filler in any matrix. Also, the product tends to have differing composition at the chill surfaces to the interior.

It was found that with this type of heating process a composition of about 40 to 50% ferric oxide could be obtained.

In order to produce greater magnetic properties, iron can be added to the ferrosilicate melt. In this way it has been found possible to increase the quantity of ferric oxide to 70%.

Other materials, including magnetic materials may be added to increase magnetic strength or provide different magnetic properties, for example neodymium boron ferrite or strontium ferrite.

An alternative heating method is to heat the powder in a process where the powder is loose, for example by letting powder pass through a gas curtain. Under these conditions it is possible to obtain powder as the end product and it is not necessary to use the high temperatures of a bulk melt.

Conversion into the ferric oxide within the slag powder will occur for example at about 9000C. The process is not clear, but in the loose form there is possibly more easily available oxygen or there may be additional surface effects due to less dense packing of the particles.

The converted ferrosilicate powder also exhibits strength and high percentage filler capability, but is additionally magnetic. When incorporated in an unmagnetised state this provides materials such as coverings, tiles etc which can be used for electromagnetic screening purposes for equipment or buildings. Such screening materials may be of any suitable matrix including glass or plastics as well as ceramics or bricks.

The magnetic powder may also be permanently magnetised and incorporated into matrices for a variety of purposes such as lightweight motor magnets. Specific magnetic patterns may alsq be imposed.

Claims (10)

1. A method of producing glass ceramic from ferrosilicate powder comprising heating the powder to a glass forming temperature in the presence of at least one additive to reduce liquidus temperature.

2. A method according to claim 1 in which the additive comprises glass.

3. A method according to claim 1 or claim 2 in which the additive comprises Sodium Oxide or Sodium Carbonate.

4. A method according to any preceding claim in which the additives are glass and Sodium Carbonate, each present inconcentrations up to 30% by weight.

5. A method according to claim 1 in which the additives are glass and Sodium Carbonate, each present in concentrations up to 20% by weight.

6. A method according to any preceding claim in which after cooling the ceramic comprises up to 80% magnetite particles in a glass matrix.

7. A method of modifying a synthetic plastic, comprising incorporating a high proportion of ferrosilicate' slag into the synthetic plastic.

8. A method according to claim 7 wherein the proportion is greater than 50% by weight or volume.

9.A A method according to claim 7 or claim 8 wherein the slag is heated to convert ferrous oxide within the slag partially to ferric oxide.

10. A ceramic or synthetic plastic made by the method of any foregoing claim.