EP1569878A1

EP1569878A1 - Method for manufacturing a ceramic article and article formed thereby

Info

Publication number: EP1569878A1
Application number: EP03735848A
Authority: EP
Inventors: David Cartlidge; Peter Samuel Hancock
Original assignee: Ceram Technology Ltd
Current assignee: Ceram Technology Ltd
Priority date: 2002-06-28
Filing date: 2003-06-30
Publication date: 2005-09-07
Also published as: GB0215037D0; WO2004002919A1; AU2003236928A1

Abstract

The invention relates to a method for manufacturing a ceramic article and to an article formed thereby. Such articles include items such as sinks, baths, and shower trays. Traditionally the manufacture of sanitaryware has been an energy intensive process. Increasing commercial and environmental demands are drivers to reduce energy requirements. It is also desirable to reduce the generation of greenhouse gasses, such as carbon dioxide (CO2) and carbon monoxide (CO). The invention overcomes problems associated with prior art processes by providing a method for manufacturing a ceramic article comprising the steps of: producing a slip including filler and clay, and a fluxing agent; forming the slip into pieces; allowing the pieces to dewater; ands firing said pieces so as to produce finished ceramic articles, characterised in that the fluxing agent includes milled glass derived from recovered glass containers. The use of milled glass has been found to reduce the temperature at which articles have to be fired. This reduces energy costs and lowers emissions.

Description

Method for manufacturing a Ceramic Article and Article formed thereby

Background

This invention relates to a method for manufacturing a ceramic article and to an article formed thereby.

The invention is particularly, though not exclusively, capable of providing a product for the manufacture of so called ceramic "sanitaryware" articles, which are items such as sinks, baths, shower trays, shower enclosures, lavatories, WC cisterns and other whiteware.

Significant difficulties are encountered during the manufacture of vitreous china sanitaryware in that natural, imperfectly understood and inherently variable raw materials are used. Sanitaryware products are vulnerable to damage as they are large, heavy, usually of complex shape and are quite fragile before firing.

An example of an existing ceramic manufacturing process is now described. The first stage in the process is to prepare a body slip, which is a dispersion of largely inorganic mineral materials in water. This is normally a two-stage operation as the materials have conflicting requirements. Ball clays need high shear treatment to achieve full dispersion. Therefore a fast-running mixer (blunger), with a rotor/stator action, is generally used. This stage can be avoided by the use of "slurry" clays, such as china clays and more commonly, slurry ball clays that are supplied ready dispersed in water. High-shear treatment of china clays degrades their properties, so a gentler dispersion action is used at this stage. Fluxes and silica can also be dispersed in this operation.

Sanitaryware is produced by some form of slip casting. Traditionally, slip is poured into a plaster of Paris mould to form the shape of the article. The plaster draws water from the slip by capillary action, causing a layer of solids to be deposited on the face of the mould as the water is removed. The rheology of the overall body composition for casting must be readily controllable and stable, that is consistent over a reasonable period of time.

Slip is allowed to stand in the mould until the required thickness of cast (typically 8 mm) has been formed. Excess slip is then drained from the mould. The cast is then left undisturbed in the mould for a period during which further water is drawn from the cast and it becomes firm enough to handle. The cast also contracts slightly during this phase enabling it to be released from the mould.

Several finishing operations are required after removal of casts from the moulds. Firstly any holes or slots, such as tap holes, in the cast have to be cut out. Secondly, more complex items, which are cast separately, are bonded together while still wet using a special high viscosity slip.

Finally, the edges of pieces are trimmed and surface pieces are smoothed. Smoothing may be done when the pieces are dry by a sanding operation, but in some jurisdictions this is not permitted as large amounts of potentially inhalable dust (which must be controlled within prescribed legislative limits) can be generated. Alternatively surfaces are sponged smooth whilst still wet. This requires a greater degree of skill, but is held to give superior finish.

Unevenness in drying can give rise to stresses that can produce cracks in the pieces if excessive. This arises principally during the first of two distinct phases. In the first phase, water fills all of the pore space in the cast and separates solid particles. During this phase the cast is still quite soft and any reduction in moisture content results in shrinkage as the particles are drawn closer together. Secondly when the moisture content is reduced to around 16% -18% (the critical moisture content), solid particles are no longer separated and the cast becomes rigid.

Cracking faults arise as rapidly drying areas contract during early stages of drying and the entire piece deforms to accommodate size changes. The damaging stresses are developed when the slower drying sections complete their shrinkage. This local shrinkage tends to pull the piece back into its original shape, but the remainder of the piece is now too rigid to allow further deformation, creating tensile stresses. Complex shaped pieces may only be dried safely over 3-4 days.

It is normal to inspect casts after drying to detect defects such as cracks, warpage or air bubbles. Some limited repairs are possible, but the majority of substandard pieces are reblunged into slip for reuse. Unfired losses can amount to 5% of the total number of pieces cast.

Sanitaryware produced since the 1950's is unusual among whiteware products in that it is only intended to be fired once. In contrast tableware is generally fired at least twice, the initial "biscuit" firing maturing the body and the second "glost" firing maturing the glaze.

The glaze slip is applied to the unfired body using atomising spray guns. Normally this produces a fired layer about 0.4 mm thick. Thicker coats of up to 1 mm fired thickness can be used to cover minor defects and body colour. Traditionally spraying has been a skilled manual task, with a sprayer taking about a minute to process a single piece, giving it two coats of glaze. Sometimes robot sprayers are employed to coat glaze.

The vast majority of sanitaryware is currently fired on cars in tunnel kilns. These are generally gas-fired kilns with direct-operating burners and run with an oxygen level of 8-10%.

The maximum temperature for most kilns is 1200 - 1250 °C. As significant amounts of water are introduced in the glazing operation it is normal to allow a short drying period before starting to fire. In many cases this is done by placing cast pieces on kiln cars, which are still warm from their previous passage through the kiln. It is essential that the pieces are virtually dry before entering the kiln as any significant amounts of water can result in the pieces exploding when heated due to steam being unable to escape from the body. One exploding piece can destroy 20-30 others through mechanical damage or gross surface contamination by debris.

Currently most UK firing cycles are between 12-14 hours. Many overseas factories run faster cycle times, but suffer higher levels of irreparable faults such as glaze pitting as a consequence. The thickness of cast sections and their low body permeability, prevent rapid burnout of impurities such as sulphates on very rapid cycles.

Fired pieces have to be inspected and graded immediately after leaving a kiln as some faults can make the pieces dangerous to handle. Typically around 5% of the pieces are scrapped at this stage and around 20-30% require some form of repair.

Many cracks do not become apparent until the piece is fired. In many cases they are present before firing but are too fine to be visible or they can run from a defect inside the piece.

Dunts are cracks developing after the body has fused on the cooling part, known as downward part of the firing schedule. They can develop from clay cracks or occur on cooling due to volume changes associated phase changes present in one of the mineral components (e.g. quartz). The glaze edges of these features are extremely sharp and an affected piece can collapse when handled, making this a very dangerous fault for handlers.

Warpage can occur due to poor handling when wet and soft or due to uneven drying. Some degree of deformation always occurs due to pyroplasticity; this has to be taken into account when the piece is designed. Spangling is the term applied to the presence of numerous small pits in the glaze surface, caused by bubbles in the glaze surface. There is no effective repair technique. Blisters are caused by gas evolution if the body is over fired. There are two possible methods of repairing minor defects. If the defect is a small crack or pinhole in a non-critical location, such as the rear of a WC closet, it can be sealed with a UV or epoxy resin that is matched to the glaze colour. If the fault lies in a visible area it is more usual to grind out the fault, stop up any resulting hole with a body/glaze mixture and to spray over the affected area with glaze. The piece is then re-fired.

In most of Europe sanitaryware is repaired by retiring to a lower temperature than the first firing, typically 1120°C - 1150°C. This minimises the likelihood that the pieces will become over fired or develop excessive deformation. Special fritted glazes are generally used to ensure maturation at these lower temperatures. However, care needs to be taken when using subsequent glazes so as to avoid devitrification in first fired glazes.

Fired pieces are much more sensitive to thermal shock on heating than in the first firing, so refire heating schedules are slower. In many cases intermittent kilns are used for this purpose to allow for these slower schedules without excessive kiln size. Despite these measures losses, particularly due to heating dunts, are common. Overall refire losses are generally 10-15%.

A disadvantage with the aforementioned process is that it consumes vast quantities of energy in order to make, dry and fire the ceramic articles.

It can be seen that the manufacturing process comprises several stages, with some running in parallel. Close control has to be maintained over each of these stages otherwise the entire system breaks down. Loss of process control is not immediately apparent, as the consequences tend not to appear until a later stage. Process control and product identification can be performed with a special barcode that survives all process steps and can be read at any point throughout production.

Any changes to materials, procedures or equipment have to be considered as part of the total production process as all operations have to be mutually compatible and practical within an operating timescale of a production plant. The most efficient plants currently achieve saleable yields of over 90% on the basis of the number of good pieces cast. More usually the yield level is around 85%. Significant variations in yield levels result from different marketing strategies; the smaller "niche" market manufacturers only achieve 70% yields or less due to the high inspection standards required.

Prior Art

Tunnel kilns are usually used in the mass production of ceramic articles. Tunnel kilns may be in excess of 50 metres long and consume Tera Joules every day. The temperature profile varies through the kiln in a controlled way in order to provide a controlled rate of heating and cooling. One reason why this is done is due to the thermal properties of the material: namely it cannot be fired (or cooled) too quickly otherwise articles could undergo thermal shocks and may fracture.

Ceramic articles are loaded onto kiln-cars, which then pass through the kiln. Close control needs to be exercised over the firing process in terms of temperature and airflow. The vast majority of kilns are fired by natural gas.

Traditionally the manufacture of ceramic sanitaryware and other ceramic articles, such as pottery, chinaware and specialist ceramic items, has been an energy intensive process. Increasing commercial and environmental demands are drivers to reduce energy requirements. The tendency to try to reduce the firing temperatures of kilns, used to make ceramic articles, is also driven with the aim of trying to reduce emissions of toxic and so-called greenhouse gasses. Increasingly primary manufacturers are being penalised, in the form of taxes, for the production of greenhouse gasses and for processes, which are energy intensive. Collectively these taxes are known as carbon taxes.

It is therefore desirable to reduce the amount of energy used in the firing process and also to reduce generation of greenhouse gasses, such as carbon dioxide (CO₂> and carbon monoxide (CO), thereby reducing so-called carbon taxes that may be levied.

An aim of the present invention is to overcome the aforementioned problems.

Another aim of the invention is to develop a vitreous china body or bodies that vitrifies at lower firing temperatures than current bodies. A further aim is to lower raw material costs.

Summary of the Invention

According to a first aspect of the present invention there is provided a method for manufacturing a ceramic article comprising the steps of: producing a slip by mixing at least 70%, by weight, of a mixture of filler and milled clay, with 30% by weight, or less, of fluxing agent; forming said slip into pieces; allowing the pieces to dewater, for example, by use of a porous mould material; and firing said pieces so as to produce finished ceramic articles, characterised in that the fluxing agent includes milled glass derived from recovered glass containers.

Fillers may include materials such materials as quartz.

An example of a typical fluxing agent is feldspar or nepheline syenite (alkali alumino silicate minerals). In an alternative method, milled glass may be mixed with another fluxing agent as an additive or bulking agent.

The introduction of a milled glass-fluxing agent has been found to reduce significantly the energy requirements for firing and as a consequence reduce emissions to the atmosphere. Use of the milled glass as a fluxing agent therefore reduces carbon taxes that otherwise would have been due. The addition of crushed glass as a fluxing agent has been found to be particularly effective at reducing the firing temperature up to 50 °C. This substantial reduction in firing temperature has resulted in considerable savings in energy costs, for example as shown in Table 1 below. It has also been found that less flux is needed when crushed glass is employed as the flux. This allows a further saving of costs because cheaper materials can be used because a higher percentage of cheaper materials like quartz can be accommodated in the body.

Waste recycling companies have been seeking outlets for reclaimed glass for some time. Recycled container glass is quite consistent in terms of viscosity vs temperature behaviour, comparing well with good quality mineral raw materials, and it is relatively rich in alkali elements like sodium or potassium -. the elements that lower melting, firing temperatures. Tests have indicated that recovered glass can be used as a suitable substitute for, or additive to, expensive mineral fluxes eg feldspar or nepheline syenite, in the manufacture of ceramic articles. Therefore a cheap replacement to mineral is effectively provided.

Additionally it has been found that when milled to preferred particles sizes and blended to the correct proportions, recovered glass flux reduces significantly the amount of energy required to fire the ceramic articles. For example ground waste glass can be blended with smaller amounts of existing fluxes such as nepheline syenite.

Preferably bodies into which the clay and flux mix are formed, are suitable to be slip cast so that they are compatible with current production moulds and production systems. Thus ideally any shrinkage of a body is similar to that of currently produced bodies. Most preferably any shrinkage of the body is less than + 1% variation of currently produced bodies.

Firing deformation is advantageously as similar to, or possibly slightly less than, currently produced bodies. A limiting value for this is believed to be in the range 80 - 100% of current deformation. An unexpected benefit of the invention is that less hydrogen fluoride gas is emitted. Hydrogen fluoride is a harmful gas. Hydrogen fluoride, and other hydrogen halides, is emitted as a result of firing the clay. As the temperature is now lower, due to the fluxing agent, less hydrogen halides are emitted.

It is important that extensive mixing of the fluxing agent is performed so that the fluxing agent is mixed intimately and dispersed throughout the clay and quartz, so as to obtain an homogenised mixture. A deflocculating agent may be used in order to assist and/or promote this mixing. Adjustment of rheology may be achieved by inclusion of acetic acid or by substitution of calcium, magnesium or aluminium sulphate, or other flocculating substances.

Ideally the percentage weight of clay lies in the range between 50% to 60%, with a corresponding percentage weight of fluxing agent in the range from 30% to 5% and sand in the range of 10-35%.

Preferably the fluxing agent includes crushed coloured glass, comprising any mixture of flint, green, brown and other coloured glass. The glass is preferably glass that has been milled or ground to an average particle size of 100 μm or less. Most preferably the average particle size is 75 μm or less.

An unexpected benefit of using crushed glass is that the ceramic article can be fired in less time than was previously possible. Thus for particular ceramic articles, such as pottery and chinaware a faster throughput is achievable. However, scheduling of other processes like casting will need to be changed.

The addition of crushed glass as a fluxing agent has also been found to improve the control of the firing of the ceramic article, by making it more predictable.

Another unexpected advantage with the invention therefore is that the previous practice of constantly having to reset and change firing times and conditions, due to variations in content of fluxing agent, is no longer required. This saves time and money and provides a more consistent quality in end product.

A further advantage of the method of manufacturing a ceramic article, is that it has been found that there is reduction in emissions of gasses such as Hydrogen Fluorides (HF), when the method is compared with existing manufacturing processes. The reduced emissions may be a corollary of firing at lower kiln temperatures and/or over a shorter firing cycle.

According to a second aspect of this invention there is provided a mixture suitable for producing a ceramic article comprising at least 80% of a sand and clay mixture and 20% or less of fluxing agent, characterised in that the flux agent comprises a mixture of conventional flux and milled glass.

According to another aspect of the invention there is provided a mixture suitable for producing a ceramic article comprising: at least 90%, by weight, of clay and milled sand and 10% by weight, or less, of recovered glass fluxing agent.

A particularly advantageous embodiment has been found to comprise at least 90%, by weight, of milled clay and 10% by weight, or less, of fluxing agent, including crushed glass.

Brief Description of the Drawing

A preferred embodiment of the invention will now be described, with reference to the Figures, in which:

Figures 1-4 are graphs showing results of apparent porosities in various trials using different mixtures of materials; and

Figure 5 is a graph showing firing results of different mixtures of materials. Detailed Description of Preferred Embodiments

Several possible methods of reducing the firing temperature of a sanitaryware body have been investigated with the aim of reducing energy costs.

Referring to Figures 1-5, trials have indicated that the use of recycled glass and feldspathic sands (both of which increase the alkali content of the body) are effective in reducing firing temperature while retaining useable firing temperature ranges. The use of the effect is limited by an increase in firing deformation if very high alkali contents are produced. Spodumene/feldspar fluxes are also effective at reducing firing temperature, although no increase in alkali content is produced.

Estimates of the effect of reducing the firing temperature on energy consumption were made. The estimates were based on a model of a sanitaryware plant producing 1 million pieces per year, with the body having a firing temperature of 1200C. Estimates are shown in Table 1.

Table 1 Estimated Energy Saving From Reduced Firing Temperature

The values in Table 1 show that the savings become significant for firing temperature reductions of approximately 50 °C. Greater energy savings are in theory possible with greater reductions in temperature, but it is believed that these would be negated by a need to use fritted glazes at greatly lower temperatures. Also, body formulation fired at lower temperature may be more prone to pyroplastic deformation. Therefore a 50 °C reduction in temperature was adopted as a target. This is because lower firing temperatures of glazes can give rise to inferior in-service performance particularly with regard to chemical and physical resistance.

A study of available data indicated possibilities in the use of a combination of spodumene and feldspar. In previous trials this type of body had required firing temperatures approximately 40°C lower than an equivalent composition containing the same total flux quantity solely as feldspar. A ratio of approximately 3 parts soda feldspar to 1 part of spodumene was indicated as producing the most effective flux.

Trials in the use of wollastonite additions to sanitaryware bodies had indicated a small reduction in firing temperature. This was considered together with data on glaze formulations that had indicated a minimum firing temperature if feldspar and wollastonite were used in proportions 1:1 to 2:1 respectively indicating that these should also be tested.

Sands that contain small quantities of feldspars, as impurities are known to have a significant fluxing effect on bodies they are incorporated in. The use of more feldspathic sands was investigated with a view to the use of kaolin processing waste.

Chemical analyses for most of the flux and silica materials used in the body samples are shown in Table 2.

All of the trial bodies were based on the following composition, taken as a generic example of a feldspar-fluxed body.

Sanblend 90 (Trade Mark) slurry ball clay 25%

Remblend (Trade Mark) china clay 25%

Sand / quartz 25%

Norfloat (Trade Mark) soda feldspar 25% Sufficient quantities of each of the materials were taken from production batches.

Samples of the bodies were made up as slips at approximately 1820 g/l, with fluidity values of 300 - 310 degrees and 1 -minute thixotropy values of 30-35 degrees. Most samples were produced by slip casting in plaster of Paris moulds. Samples for firing deformation trials were formed by extruding slip that had been flocculated and dewatered to produce a plastic body.

Each body was fired in a gradient kiln then tested to determine the temperature required to achieve vitrification with a water absorption of 0.5% or less (T1) and that required to achieve maximum fired bulk density (T2). The difference between these values provides an indication of the usable firing range of the body. Firing deformation was determined by using the Pyroplastic Index method. Samples were fired to T1 and T2 for this purpose.

Four types of silica were used in the trials:

1. Sibelco (Trade Mark) HPF2, a high purity Oakamoor (Trade Mark) quartz.

2. Sibelco (Trade Mark) Chelford sand contains some feldspar and mica.

3. CMO 305, a feldspathic sand provided by English China Clay (ECC) as an example of the waste material from china clay production.

4. Feldspar Corporation Sil-o-Spar (Trade Mark), commercially available feldspathic sand from the USA, used by some sanitaryware manufacturers.

Comparison trials of the two Sibelco (Trade Mark) materials in the base body showed the quartz variant required the highest temperature to achieve vitrification approximately 1220°C and reached maximum bulk density at approximately 1250°C. The Chelford sand variant achieved vitrification 5-10 °C lower. The CMO 305 sand produced a marked reduction in firing temperature, achieving vitrification at 1150 °C without any marked reduction in firing range. Peak bulk density occurred at 1190 °C.

The Sil-o-Spar (Trade mark) was used as the supplier recommended, as the sole non-clay component in the body. This again gave a significant reduction in firing temperature, representing a 40 °C reduction in firing temperature over the "base" bodies. Again the firing range remained at approximately 40 °C. Firing properties of these bodies are illustrated in Figure 1.

A powder grade of wollastonite, ground to pass a 75 μm mesh was used. The initial trial bodies (W1 and W2) used feldspar/wollastonite blends as described above, with two levels of substitution of wollastonite for feldspar used. These changes from the base composition actually degraded the fluxing action, as shown in Figure 3, indicating that the glaze data did not reproduce in bodies.

Additional trials were also conducted in which the feldspar content was kept at around 25% by weight of the base body and wollastonite substituted for some of the silica component. This also proved unsuccessful, with the vitrification temperature being higher than the corresponding base bodies.

Spodumene / Feldspar Fluxes

In these trials "Universal" grade spodumene, supplied by Sons of Gwalia, and milled to 200 μm mesh was substituted for some of the feldspar in the base body. Both quartz and feldspar variants of these were tested.

The results from several trials of this type of body modification are shown in Figure 4. Initial trials using a 4-6% substitution indicated a reduction in firing temperature. However, this was less than seen in previous trials at approximately 20 °C below the base body temperature. As the 6% substitutions gave very similar results to the 4% equivalents, the results indicated that there was a range of compositions over which the effect was stable. Further increases in spodumene substitution resulted in reduced firing temperature; a combination of 10% spodumene and 15% feldspar gave a 40 °C reduction.

Recycled Glass / Feldspar Fluxes

Samples of recycled container glass ground to pass a 75 μm sieve were used. The ground glass comprised a typical blend of green, clear and brown glass that is obtained from most waste recovery operations. The glass was used as a partial substitute for the feldspar in the base body. However, as colourless glass is more expensive than coloured (green, brown or blue) glass, it is envisaged that only coloured glass will be used in practice. Coloured glass is suitable for this, since an opaque glaze is used in sanitaryware and this masks any discoloration in the fired body that comprises a coloured glass flux. Viscosity measurements and calculations both show consistent η vs T profiles regardless of both glass mix. That is green, brown and clear glass behave substantially the same as one each other.

Using equal portions of glass and feldspar, a reduction of 40-50°C in the firing temperature compared to the base body was produced, as shown in Figure 5. The firing range was still comparable to the base body at 30°C.

The effect of the particle size of the glass was tested by further grinding of the glass samples provided. A sample ground to pass a 45 μm sieve was prepared and tested in the body. This did not produce any significant reduction in firing temperature, since the reduction in porosity as peak temperature approached was more rapid, a reduction in usable firing range could be expected.

It was noted that the glass had a deflocculating effect on the slips that contained it; this means the suspensions became more fluid or lower in viscosity in time. In some cases a fine adjustment of fluid properties by a small addition of plaster (0.01-0.03% weight) is necessary. Alternatively, inorganic compounding capable of trapping the alkali ions leached from the glass powder can be used.

Wet milling of glass can be used to leach out unwanted alkali ions. A silicon rich surface layer results such that when milled glass is dried and then re- dispersed in water slip stability results. It was also noted that the bodies containing the glass had a deeper colour than normal. In practice however, this darker body colour would not have any detrimental effect to the finished product because the opaque glaze used provides the visible colour in sanitaryware.

To determine whether any undue changes in firing deformation had been produced, selected bodies showing promise for further development were tested in comparison with the base bodies.

The results from these tests are shown in Table 3. They show no significant differences between the two base body variants. If results representing comparable points in firing are compared the Sil-o-Spar (Trade Mark) body shows slightly less deformation than normal.

The glass body shows slightly less deformation at T1, but slightly more at T2. This indicates a slightly narrower firing range has been produced.

The body containing the CMO 305 sand showed a general marked increase in deformation, both at T1 and T2, indicating that no body type would be unsuitable unless some additional technique was employed to reduce deformation.

All of the bodies tested, apart from the wollastonite types show very similar shrinkage values, indicating the no difficulties should arise from this.

Therefore from the aforementioned trials, recycled ground glass flux provided an effective fluxing action. It is a low-cost raw material; and has been shown to be effective in reducing firing temperature. To ensure that the essential working properties were not being degraded in an attempt to reduce firing temperatures, slip samples of the "base" body and the 12.5% feldspar + 12.5% ground glass flux body were prepared. These were set initially to 36.6oz/pt with Gallenkamp (TTV) fluidity and thixotropy values of 310°C and 30°C respectively. Both slips showed self-deflocculating behaviour during ageing. The thixotropy of the "base" body reduced to 18 degrees while that of the glass/feldspar body fell to 8 degrees over 1 week. This is consistent with observations that glass-containing bodies were self- deflocculating.

Tests were made of viscosity development (before ageing changed the state of deflocculation), cast formation rate, cast moisture, cast firmness, shrinkage, Modulus of Rupture and firing deformation on the two samples.

The two slips showed very similar viscosity development, with values between 7500 and 8000 MPa being reached after 60 minutes standing.

Tables 4 and 5 list the casting properties of these two body samples. Some differences were noted, in particular faster casting and firmer casts and slightly greater shrinkage from the glass/feldspar body. The firing deformation test was carried out using an industrial "sag bar" test run at T1. This showed the glass/feldspar-fluxed body to give less deformation than the base at this temperature, as in the previous Pyroplastic Index tests.

Overall, the working properties were considered to be within the useable range, although there was concern over the self-deflocculating nature of the glass-containing bodies.

A further investigation was made of the possibility of using greater proportions of glass. In this investigation bodies that were fluxed solely by glass were tested. Some modification of the body composition was required to achieve the same vitrification temperature achieved in previous trials. The 40-50°C reduction from the vitrification temperature from the base body was maintained to permit direct comparison.

The required firing temperature was achieved with a composition of:

25% china clay 25% ball clay 32% Chelford sand 18% recycled glass

Figure 6 shows the response to firing temperature and indicates a potential drawback in that the useable firing range is reduced in this body. Table 6 shows that the deformation values for this body were also more sensitive to firing temperature, being lower than the "base" body at T1 but higher at T2. This behaviour appears to be related to the amount of glass used in the body.

Some changes would have to be made to this body to make it suitable for use in an industrial situation. Possible measures include manipulation of the particle size of the glass and/or use of a small amount of a supplementary flux.

The use of feldspathic sands and glass can both be regarded as using an increased proportion of alkalis in the body. The glass probably has a secondary effect in that it has already received significant heat treatment and can be regarded in a similar manner to adding a frit to a glaze.

There appears to be a limit to how much the alkali content can be increased in practice and the body appears to represent this, due to its high firing deformation. Also thermal expansion of the body can increase leading to firing cracks and glaze thermal expansion mis-match.

The said firing cracks are known as dunting. Overall, it appears to be possible to produce bodies by this route that can vitrify at approximately 1160 °C (T1), have a firing range of approximately 40 °C and retain similar deformation values to current types. This would represent a reduction in firing temperature of 50-60 °C over the base bodies that have been used. There is an optimum amount of quartz and fluxing agent because of phase changes that occur between x and β phases. These phase changes can give rise to stress cracking and therefore, although in theory larger amounts of flux can be added, there is a practical limitation imposed by stresses caused by phase changes.

The spodumene/feldspar fluxes were effective in reducing firing temperature without an increase in alkali content, provided a sufficient proportion was used. This type of flux system has previously been shown to reduce firing deformation at moderate levels of substitution. This behaviour could either be employed alone or in combination with a markedly increased alkali content flux system to restore deformation to normal levels.

Recycled glass has proved to be effective and appears to have advantages in raw materials cost.

Table 2

Chemical Analyses of Materials Used

Table 3

Pyroplastic Deformation Values for Selected Body Samples

Table 4 Casting Properties of Trial Bodies

Table 5 Comparison of Strength and Deformation of Trial Bodies

Table 6 Firing Properties of Glass-Fluxed Bodies

The finely milled glass can be added into the clay processing stream of the clay at a number of points. The nature of where in the clay processing stream the glass is added is reflected in the manufacture and method of addition as identified below:

An advantage of this method is that the material can be easily handled at the factory with no extra capital needed. On simple plants the tempered powder can be handled by a front-end loader. As the colour of the body of the article is of low concern, because all that is seen in the fired product is a white opaque surface glaze, full advantage of the properties of the replacement feldspar can be taken. This may also be applicable to coloured glass.

The invention has been described by way of exemplary embodiments only. It will be appreciated that variation to the embodiments described herein may be made, without departing from the scope of the invention. For example, although the invention has been described with reference to compositions containing UK china clays and ball clays, it will be appreciated that other mixtures of clay would be suitable. Similarly a myriad blend of glass with existing fluxes is envisaged.

Claims

1. A method for manufacturing a ceramic article comprising the steps of: producing a slip by mixing at least 70%, by weight, of a mixture of filler and clay, with 30% by weight, or less, of fluxing agent; forming said slip into pieces; allowing the pieces to dewater; and firing said pieces so as to produce finished ceramic articles, characterised in that the fluxing agent includes milled glass derived from recovered glass containers.

2. A method for manufacturing a ceramic article according to claim 1 wherein the percentage weight of the milled sand and clay lies in the range between 70% to 95%, with a corresponding percentage weight of fluxing agent in the range between 30% to 5%.

3. A method for manufacturing a ceramic article according to claim 1 or 2 wherein the fluxing agent is mixed intimately and dispersed throughout the clay, so as to obtain an homogenised mixture

4. A method for manufacturing a ceramic article according to any of claims 1 to 3 wherein the fluxing agent includes crushed coloured glass.

5. A method for manufacturing a ceramic article according to claim 4 wherein the crushed glass has been milled or ground to an average particle size of 100 μm or less

6. A method for manufacturing a ceramic article according to claim 4 or 5 wherein the crushed glass has been milled or ground to an average particle size of 75 μm or less.

7. A method for manufacturing a ceramic article according to any preceding claim wherein additives are added to the wet body mixture in order to remove unwanted substances, such as alkali elements such as sodium, potassium and lithium.

8. A method for manufacturing a ceramic article according to any of claims 4 to 6 wherein the fluxing agent includes crushed glass wherein means is provided for removal of excess sodium.

9. A method for manufacturing a ceramic article according to claim 8 wherein the means includes chemical absorbents such as zeolites or electro-chemical processors.

10. A method for manufacturing a ceramic article according to any preceding claim wherein a deflocculating agent is added in order to promote mixing of the clay and fluxing agent and promote a stable suspension.

11. A ceramic article manufactured according to any of claims 1 to 10 wherein the addition of crushed glass fluxing agent has been found to improve certain physical properties of the ceramic article.

12. A ceramic article manufactured according to claim 11 wherein the frost resistance of the ceramic or similar article is superior to existing ceramic or similar articles.

13. A mixture suitable for producing a ceramic article comprising: at least 70%, by weight, of milled sand and clay and 30% by weight, or less, of fluxing agent.

14. A method for manufacturing a ceramic article substantially as herein described with reference to the Figures.

15. A ceramic article substantially as herein described with reference to the Figures.

16. A mixture suitable for producing a ceramic article substantially as herein described with reference to the Figures.