IE45447B1 - Improvements relating to asbestos-free fibre reinforced cementitious products - Google Patents

Abstract

Cementitious prodts. reinforced with fibres are mfd., in which the mixt. is formed into a short from a fluid suspension by applying a suspension of cement and chopped glass fibre rovings to a permeable surface and draining off the liquid by suction through the surface. Specifically a flocculant is added to the mixt. together with bon-crystalline inorganic monofilaments esp. of glass wool or mineral wool. The prodn. replace prior art asbestos cement prods. The flocculant additive improves the drainage, so that the water content of the setting cement is reduced, Giving increased strength to the finished prod. The fine monofilaments prevent fine particles of filler and cement from passing through the drainage holes, and provide in their own right increased reinforcement.

Description

This invention relates to asbestos-free fibre-reinforced cementitious products, and more specifically to methods of manufacturing asbestos-fred fibre-reinforced cement composite ί materials in which the reinforcement is provided by glass fibres, alone or in conjunction with other fibrous substances.

In the production of .asbestos-cement products, it has been customary to deposit a water-laid web of cement and asbestos fibre on a foraminous (i.e. perforated or porous) surface from a fluid slurry of these materials, and to de-water the product by suction and/or pressure, ' Equipment has been designed and is widely available for carrying out this method on a large scale, hut a need has now been recognised to -replace or considerably reduce ; the quantity of asbestos used in such products. • 15 The equipment at present used to manufacture asbestoscement products is designed to take advantage of the ability of asbestos fibres to flocculate, and to be wetted by, an aqueous cement slurry. Two principal types of machine are used, both known by the name of the original inventor. They are the HATSCHEK and MAGNANI machines.

The Hatschek machine produces a sheet composite material from a thin aqueous slurry of cement and asbestos fibres by a step in which a water-laid web of these materials is deposited on a foraminous surface and de-watered by suo25 tion. The foraminous surface is a rotating cylindrical sieve.

It is important to avoid substantial losses of cement due to fine material passing through the foraminous surface. The . slurry or furnish, as it is called, must therefore be in a state in which the filtration rate is not so slow as to ¢5447 make production uneconomic, but in which the filtration rate is not fast merely because the quantity of fine material which passes through is high.

Similar problems are encountered ixi operating a Kagnani machine even though the slurry in this case is more dense and not so fluid as in the case of the Hatschek machine. In the Magnani machine, a continuous travelling belt of felt clorh is supported by a perforated base through which suction is applied to the underside of the belt, while jo a dense but flowable slurry of water and cement containing asbestos fibres is applied to the upper side of the belt by means of a reciprocating distributor which travels back and forth above the belt and is fed with the slurry from a storage tank whose contents are continuously agitated by a mechanical mixer. The distributor moves faster than the travelling belt and thus builds up a sheet of asbestos cement on the belt in thin incremental layers, which are de-watered by the suction applied to the underside of the belt. Other known machines for making asbestos-cement pipes operate on very similar principles.

The characteristic ability of asbestos fibres to flocculate and to be wetted by the cement slurry or furnish generally enables the slurry to be formed in a suitable state to achieve the desired de-watering rate without excessive loss of fines, in the known machines referred to above, though various additives are often also incorporated in the slurry or furnish to control the de-watering or separation rate. If the asbestos fibre is simply replaced by glass fibre in chopped-strand form, as a reinforcing material, the results are not very satisfactory due to the fact that the - 3 glass fibre has no substantial flocculating action and is not wetted by the cement slurry, with the result that excessive quantities of fine materials pass through the foraminous or water-permeable surface, i e. the cylindrical sieve of a Hatschek machine or the felt belt of a Magnani machine. A consequence of the loss of these fine materials is that the product lacks green strength and tends to delaminate.

Glass fibre is available in two principal forms, namely continuous filament, which may be· combined .into strands and chopped to specified lengths, and non-continuous single filament. The main division between these available forms of glass fibres is based on both the process and equipment used for their manufacture and the form in which they are produced. Glass fibre in continuous filament form is made by drawing single filaments from minute streams of molten glass issuing from orifices in the base of a container known as a bushing. The filaments are sized immediately after they are drawn and gathered into groups of filaments which are known as strands. Such strands may be chopped 20 to provide discrete bundles of filaments arranged in a linear form and bonded together by the size. The length of the strands is determined at chopping and can range, for example, from as short as 3 mm to as long as 30 mm. The number of filaments is determined at the drawing stage and the filaments can either be gathered into one large strand or split into several strands which are wound together hut separate on chopping. The strands are wound into a cake which after drying may be used with a number of other cakes to form a roving which is a grouping together of a plurality of strands. A roving is 3θ one form in which glass fibres may be fed to a chopping gun to produce chopped strands, though strands can be chopped while still wet with the size immediately after leaving the bushing. - 4 Chopped strands produced in any of these ways are those referred to above as being used as a reinforcing material.

They are already used in reinforcing both polymeric materials and inorganic cement matrices but as indicated above have caused difficulties when used alone on incorporation in cement when operating with the type of machines used for making asbestos-cement products.

The continuous filament method is only used to manufacture fibres of glass, because of the nature cf the process. The so-called discontinuous processes can be used to make fibres from glasses, mineral sources such as rock or slag and ceramic materials. The fibres are produced in single filament form and are not grouped into bundles or strands with a substantially linear arrangement. The products include glass wool,mineral wools and steam blown filaments. One well known discontinuous process involves allowing molten glass co issue from orifices in the peripheral wall of a vessel rotating at high speed, and attenuating the streams of glass by a blast of hot gas. Single discontinuous filaments can also be produced by attenuating the streams of glass issuing from orifices in the base of a platinum bushing with a blast of steam. The older discontinuous process known as the Hager process simply involves allowing a stream of molten glass to fall on to a rapidly spinning grooved disc.

A difficulty with this latter process is that as well as single filament fibrous material, small spheres are produced, customarily known as shot”.

The single filament products of the discontinuous processes are of little value for reinforcement purposes and are normally used in forming insulating mats and the like. - 5 .. 45447 Single filament material can also be produced by adding to an aqueous medium, chopped strands of continuous t filament glass fibre which 'have been sized with an aqueous size but which have not been dried, or which have been sized with a size which after drying is still water-soluble or dispersible.

We have now found that single filaments which are inorganic and non-crystalline can be used in enabling asbestos fibres to be wholly replaced in a conventional asbestos cement manufacturing process by chopped strand glass fibre without the disadvantages explained above, so that available asbestos cement manufacturing equipment can be utilised without substantial changes being needed in the equipment or methods of operating it.

U The invention accordingly provides a method of manufacturing an asbestos-free fibre—reinforced cement composite material in which a water-laid web of cement and continuous-filament chopped glass fibre reinforcing strands is deposited on a foraminous surface from a fluid slurry of those materials, and is de-watered by suction, wherein prior to the web being laid a flocculating agent is added to the slurry and in addition to the chopped strands there are present, in the slurry, single filaments of inorganic non-crystalline material.

It is believed that the single filaments may act as effective nuclei for the flocculated cement particles, thus forming a homogeneous matrix of single filament fibre with the cement floes. Another possibility is that the single filaments are agglomerated by the flocculating agent. Whatever the mode of action may be, the. method of the invention has proved to make it possible to reduce the loss of fines to acceptable proportions. - 64-3 417 The cement composite materials thus produced have advantages from the point of view of health, because of the absence .of asbestos, and also have good impact resistance while retaining the fire resistance and other desirable properties of conventional asbestos-cement products.

The single filament material used may be a glass wool or mineral wool, or it may be produced from continuous filament chopped glass fibre strands which separate or filamentisc on contact with the cement slurry.

As a general but not universal rule the effectiveness of single filament material in retaining cement fine3 has been found to be greatest when the mean diameter of the filaments is least. Larger diameter fibres therefore generally have to be used in larger quantities to achieve similar results. Very fine glass fibres are expensive and difficult to handle and a balance may need to be drawn between cost and handling problems with fine fibres and the problems involved in using larger quantities of larger diameter material. In general, usable single filament fibres will normally have diameters in the range of 0.5/nn to 20yum, while the single filament fibres which are preferred from the point of view of cement retention have a mean diameter in the range from 4 to 9/tm. The single filaments used may be from 3 to 30 mm in length, but preferably not more than 12 mm.

Addition of an amount of the single filaments (with diameters in the range 4 to 9/im) sufficient to constitute from 2 to 6% by weight of the total solids in the slurry has been found sufficient to reduce the escape of cement fines to acceptable levels, comparable to- those experienced with asbestos cement. Coarser fibres will of course involve the - 7 use of increased quantities of fibre. As a general rule, the proportion of single filament material in the slurry will be from 1% to 10% by weight of the total solids.

The single filaments may be dispersed in water before addition of the cement and the chopped glass fibre reinforcing strands and other materials to form the slurry. Alternatively the aqueous dispersion of the single filaments may te performed in the presence of the cement during production of the cement slurry.

We prefer to use a flocculating agent which is anionic * in character and has been derived by modification of an acrylamide polymer, e„g. by forming a high molecular weight copolymer with an acrylate. We refer to such preferred flocculating agents as polyacrylamides. The polyacrylamide flocculating agent may be added in an amount to constitute from 10 to 1000 parts per million of the dry solids content of the slurry, and preferably from 50 to 600 parts per million thereof. Preferably the polyacrylamide flocculating agent is added to the slurry immediately before deposition of the web.

We prefer to use inorganic non-crystalline single filaments which either have been made by a discontinuous process in which little '’shot” is produced, or have been substantially freed of shot before use. It is also however, possible to use continuous filament chopped glass fibre strands which disperse or filamentise on contact with the cement slurry.

The cement is normally ordinary Portland cement, though any other hydraulic cement may be substituted. A portion of - 8 10 the cement may be replaced by one or more other materials in the nature of fillers, e.g. for controlling the shrinkage on drying, or the density or alkalinity of the product.

Specific embodiments of the invention will now be described by way of example.

Experiments were made in the laboratory to determine the suitability of various slurries for use in a Hatschek type asbestos-cement plant, in which a web of cement and fibre is deposited on the foraminous surface of a cylindrical sieve from a thin aqueous slurry of cement and fibres, and is de-watered by suction.

The trial batches of slurry were made up to the following general formula:15 Ordinary Portland cement Chopped-strand alkali-resistant glass fibre Single filament material Polyacrylamide flocculating agent (Aquafloc 4103 ex Dearborn Chemicals Limited) Water to give a water/cement ratio of Percent by weight 90% 6% 4% added as 0.01% aqueous solution at 250 ppm of solids 23:1 (including flocculant solution) Comparisons were made during these trials with single filament materials which have been previously used in asbestos-cement machines, namely chrysotile asbestos in slurry 1, cellulose derived from white wood pulp in slurry 6 and cellulose derived from newsprint in slurry 7. Slurries 2, 3, 4, 5, 8, 9, 10, 11 and 13 used various types of glass wool as the single filament material. Slurry 12 used mineral wool as the single filament material. The following Table 1 lists the results obtained when filtering the slurries through a 300 fua mesh sieve, simulating the effects to be expected on a Hatschek type asbestos-cement machine. ίω , η bο ri Ο Ν in s iSΟ σ0 cn oj Η Ο β Λ μ <ΰ£ μιο μ ο ΦΟ SHO ΡΡι CD H VO ts in o CM to 00 o CM <$ rH CM CM CM tn in tn to 3· tn •4* to o « • • • « • • • • • o o o O o o o o o o o o o IS cn co -ra- to MO s CM H ri CM CM in to m tn • » • 0 » • o o O o o o o o O on ω νο ΙΑ m m Ο Ο Ο. Ο σ> Η ο Ι<Ί ri ΙΑ Μ) !Π Ν 05 ΙΜ νί) ια η m < β · β · · Ο Ο Ο Ό Ο Η *Η Φ ra hi a •si ί ο ο co φ I φ I xi ο « $ « β <3 eS ΐ Ό β Ο •Η +> S •μ Η ri fa to /— a 3 ri ri a ra a ra S o μ·κ o ai+j v £ ra ra •a μ a a ra ra μ — ' ra ri riri fa Ό s o VO s CD o (S CM •a- to oo cn ri CM CM CM CM CM m m to •ra- •ra- to • • • β • • • • • • * • • o o O o o o o o o o o o o ο o m o ri cm ri m in -rais m CM MO vo ο in -ra· o in mo m t- m . in CM Π o O s to 45 • CM 6 tn St +1 • CM in y tn ra μ ri in ttn IS cm m m · o · sf * cm’ * +i +l+. +l _ cm oo m · -ra* O · CM O\ ri. Η 'Η Φ ri Ιώβ Μ Ή 'ϋ ra Ή ri Ή Ο O U ?0 Ο 3. riri <-» Μ μ ra a ri. g to ε a 3 riri Mri fa £ Ϊ Si «! ri H H H S fi X“X O O O •H ' Pl Φ P o C> o rH H to ti 2s 2= Ss 0 Φ ti O-H O xf •w ft H ti w to to JS Φ o ti ft w to to ti rH Ό H to (0 Φ <0 to β 3 O H £ h H rH to o Η O Φ Φ 0 0 0 Φ ti d £ O ti o ra ra cm m -a· tn MO CM riri o — 003 MOB ra ra ra ra tiri riri O ra · a Λ co oo ri rim ri O Z— o 003 p * ra ri a ra w ri ri ri O O O Ο O Q is ΪΒ > ri m ra ra ra n u ra u ra tj ri ra ra a ra ri ri ri ri ra ri ora ο o a o Ei ri o\ O ri CM m i-l ri ri H AS 4 47 Three runs (.A, B and C) were made with each slurry and measurements were made, as listed in Table l’, of the de-watering time, the percentage solids in ihe filtrate, and the percentage of the cement in the slurry which had thus been'lost through the sieve, ft will be seen from Table 1 that the single filament materials of the smallest diameter were the most effective in retaining the cement.

The percentage solids in the filtrate increased' fairly uniformly with fibre diameter, with the coarsest glass filaments retaining che least cement.

It can also be seen that the de-watering time increased with the diameter of the filaments to a maximum at between 5 and 6 pin, and thereafter decreased for the larger filaments. It Is possible that, with low filament diameters, increases in diameter cause a decrease in the size of the floes formed, and hence a more compact filter cake and increased de-watering time. At higher filament diameters, the dominant factor is probably the filament diameter itself so that further increases in diameter begin to increase the drainage rate and hence reduce the de-watering time.

The degree of retention of cement was found to depend upon the amount of single filament material used. For this - 11 _ 4S4 47 aspect of the investigation, three different types of glass wool with diameters of less than 2.^wn, 7.5yUm . and 12.8/im, were incorporated in slurries similar to those described above and subjected to similar filtration tests.

The results are set out in. the following Sable 2.

TABLE 2 Effect of monofilament concentration on filtrate solids and oe-waf οηηκ -time Single Filament type Single Filament diameter ^xm) level of single filament in slurry (% of other slurry solids) de-water ing time (seconds) % solids in filtrate 1. Control 0 ! 125 ·. j 0.75 ! 2. Glass Wool 12.8 2 112 i 0.54 4 50 1 0.54 i 6 45 0.36 8 33 0.30 3, Glass Wool (shredded to 12mm length) 7.5 { 2 4 δ 120 69 36 0.43 0.35 0.29 4» Glass Wool <2.4 1 23 0.21 (approx 1.5) 2 22 . 0.19 4 20 0.21 It can be seeii that a given filtrate cement content can be obtained by the use of smaller quantities of finer filaments, io A filtrate cement content of 0.3% would correspond to a loss of around 10% of the cement in the slurry which is approximately the level normally experienced with asbestos-cement on a Hatschek machine. Extrapolating from the results shown in Table 2 one could expect to obtain this performance by addition bf less than 1% of 1.5/ The quantities of single filament materials referred to above are expressed as percentage additives to the total solids in the slurry.and. they are correlated with estimated 12compositions of the final cured glass fibre reinforced cement composite material in the following Table' 3. TABLE 3 - GRC Compositions Single Fila- iEstimated % comoonents in final cured comoosite· ment material % solids in slurry Cement Single filament material ΐ 12 mm length chopped glase- | fibre strands Water 0 72.9 — 5.1 22.0 2 71.9 1.5 . ‘5.0 21.6 4 70.8 3.0 4.9 21.3 6 , 69.7 4.5 4.3 4.8 ; 21.0 8 68.6 5.9 20.7 ! —......—J * Based on a final water/cement ratio of 0.30 Flocculant! A4103 at 250 ppm (solids) Table 2 also shows that increasing the quantity of single filament material also decreased the de-watering time, and that the larger diameter filaments showed the greatest variation of de-watering times over the single filament concentration range examined. The de-watering times given by the finest single filaments were relatively insensitive to concentration. Use of 4% of the smallest diameter single j5 filament material reduced the de-watering time of the control mix by 84%.

The effect of the length of the single filaments on the retention of cement is indicated by slurries 8 and 9 in Table 1. The single filaments used in these slurries were 2o similar except that they were shredded through mesh screens to lengths of 12 mm and 31 mm respectively before the dispersion was prepared. The comparison suggests that the shorter filaments are more effective but will give longer de- 13 ¢244^ watering times« Filament length is, however, of direct relevance to preparation of the single filament dispersion, which is believed to be best accomplished by violent agitation of a suspension of the filaments in water, using a hydropulper or other high shear mixer. If the single filaments are not thoroughly dispersed, lumps of filaments are observed in the web. Shorter filaments are more easily dispersed and it is preferred, that filament length does not exceed 12 mm, though a length of up to 25 mm may he 10 sufficiently easily dispersed. The method of dispersion of the single filaments, using a high shear mixer, may be expected to inflict considerable mechanical damage on the filament surfaces and it is not expected that they will contribute substantially to the reinforcing action which is IS provided by the chopped-str'and. glass fibres.

When operating on an asbestos-cement plant, a balance must be found between use of excessively large quantities of large diameter fibres (up to 20/Xm) which could involve difficulties in handling and storage and possibly deleterious . I effects on the properties of the final composite material, and use of smaller quantities of exceedingly small diameter filaments (e.g. 0.5 /Xm) which are the most expensive and also bring handling problems. The preferred single filaments for cement retention are of a mean diameter of around 4 to 9/Wm 25 though fibres of greater diameter may be used if the increased quantities required are acceptable. Scrap glass wool Which has not been dried after application of a conventional binder has been found to be both effective and cheap.

Reverting to Table 1, as can be seen from a comparison of the results obtained with glass wool and the results with - 14 slurries 1, 6 and 7 wherein in the first case asbestos was used, and in the other two cellulose, it is possible using glass wool to approach the same level of cement passed through a 300^m mesh screen as with asbestos, and to obtain results equivalent to those obtained with cellulose. The degree of retention is,however,in all examples of the invention sufficient to enable the invention to be practised successfully on a commercial scale. It is thus possible to replace asbestos and avoid any possible health hazards arising ires its processing and use. Cellulose, while effective, poses a problem in that it increases the combustibility of the materialeven * when used in conjunction with chopped strands of glass fibre.

It can also have an adverse effect on the strength of the composite. It is possible in some cases where certain properties are desired in the finished board to utilise cellulose fibres in conjunction with chopped strands as the reinforcing material, and in addition glass or mineral wool to reduce the amount of cellulose to a level where satisfactory non-combustion properties are obtained. In such a case the amount of glass or mineral wool used is also reduced. The use of mixtures of non-continuous inorganic non-crystalline single filaments with cellulose is therefore dependent on the physical properties desired in the finished product and cost. Cellulose has a performance roughly equivalent to that of 5//m diameter glass filaments but the quantity of glass filaments required to replace a given proportion of cellulose is of course not equivalent in cost or in its effect on the solids content of the slurry. Glass and mineral wools are available commercially in the preferred range of filament diameters at reasonable cost, and so in general we prefer to use these - 15 materials in the method of the present invention.

It is, however, also possible to use continuous filament·chopped glass fibre strands which separate or filaraentise on'contact with the cement slurry, to produce the single filament material.

Where the single filament material is derived from dispersed or filamentised chopped strands, the material as normally available has a diameter in the upper' end of the range 0.5yWm to 20 /,<.m quoted above, Le. the filaments are usually not less than lOyt-tm and not more than 20yum. in diameter. This means that, if this material is used ' alone to give good cement fines retention, relatively large quantities - of fibre may be needed. We have found it advantageous to use mixtures of small diameter fibre such as glass wool in conjunction with the single filament material derived from dispersed or filamentised chopped strands. In this way we take advantage of the smaller diameter material'.^. ability to improve cement retention with relatively smaller additions of the material, along with the fine surface finish and good green strength derived from the presence of the matezu rail derived from dispersed or filamentised chopped strands.

We have found that one way.complete and' uniform filamentisation of chopped strands can be obtained is by adding the glass fibre strands in a wet and undried state to the cement slurry or furnish. The material in wet form can be obtained either by chopping the strands as they leave the bushing and packing them in containers where little or no drying takes place, or directly winding the roving or oake and wrapping the roving package or cake so that it cannot dry out.

We believe iii this case it is unnecessary to use any of the normal filo formers in the size applied to the filaments during drawing. It is not possible however to achieve a Ιβ >4·ΰ 4 4 7 satisfactory product without sizing, and we therefore use water· with a small amount- of a standard size lubricant such as TWEEN (Registereu Trade Mark) 20. U.S. Patent Specification No. 3,94^,^73 suggests suitable size compositions which when applied to glass fibre strands give a product which when chopped after drying is dispersible in a cement slurry. Care is needed in choosing such a size as many so-called dispersible sizes result in a product which when dried is difficult to get completely open. This results in bundles with an open brush like end which can become entangled with other bundles resulting in clumping of the glass and hence a poor product. We prefer when following this route to use a size which has been evaluated under operating conditions particularly where the size is one subjected to any degree of curing during drying.

In order to demonstrate the effectiveness of the use of single filament material from dispersed or filamentised chopped strands, with regard to achieving the same order of solids in the filtrate as in the situation where asbestos alone is used on a machine, we have carried out a series of filtration tests using a 300 pm sieve. This provides numerical values under standard conditions which can be compared in respect of this , >43 447 particular test. The actual figures obtained in plant runs can be variable, and are not necessarily comparable directly one ’ with another. In general we find that a level of less than 1$ solids in the filtrate on these tests indicates that a mix can be made to run effectively and economically on the plant, assuming that the problem of handling the product in a comparable manner to asbestos cement composites in the green state has also been solved. The following Table 4 illustrates that the solids ' level obtained in the presence of integral chopped strands and single filament dispersed chopped strands, with a flocculating i agent, (see Mix 7) is of the same order as with asbestos alone, ’ or as in mixes 4 and 5 with only the dispersible glass fibre and i the other additives, i.e. flocculating agent alone in mix 5, and < cellulose and flocculating agent in mix 4. Mixes 2,3 and 6 show that omission of the flocculating agent produces unacceptable t : results in the absence of asbestos. - 18 TABLE % SOLIDS ft ft X< ft ft ft ft ft ft © ft- 1 o o ft* i '© rO \C © I ff, co © 0.6-1 .0 I 4.1 © ft ft a OT 0 > ft z; ft IA to rt co tin CO Η H ft © © rt d © • <· ft OT OT cl 1 I 1 I o Γη. © Cj OT , rt Ό \O rt »“4 rt Vi «ί Λ 00 00 • rt V. a, o 'JC r\ CO ft s s € -τ' CijCC CL, CO ftco ft 0,0 ftO ft© P rt rt I—1 u } 1 ! © r* © rt 1 © rt tT) in •A 5 c N < CI ·< Cl <*. i ft ft ί ft ft ft T-l In (n. Ά Γη Γη •n < Cl • • • • • * £ ro co CO co ^O co CI CI d Cl ❖ ft ft OT ft O ft ft ft e-t O OT O H OT ft T—< I J 1 I 1 I OT ft SO © g Λ ft X < A o O CJ « ft oi ft © OT ft o·© ft ft ft X ft I rt I rt < ft X n &H H ft IX. ft ft ft OT ft e-( ft ft X « < OT ft m m ω ft OT OT ft • • 52! ft ft «ί ft I tn in in ci Cl © © W ft ft ft X a. e u, ss u OT o ft u o Vi ft w § ft ft ft OT s ft S Oh Q in m w 1 ft x • • e. g o < I V) 1 1 d d OT ft ω υ ri ft ft *-*OT ft X co co tn co co tn ft © • • • • • • £ rt rt o rt rt © 0 tn Γη tn. Γη m ft M · tig A A T-< M co rt in 'P fn gn < & 6-t x g w ft w ft X H ft .Q ft X OT OT < '1 I The flocculating agent chosen is preferably a polyacrylamide such as those sold under the trade-names AOUAFLOC ( Trade Mark) 4103 (a Dearborn Chemicals Co. product) or MAGNAFLOC ( Trade Mark) E 24 (an Allied Colloids Ltd product). The quantity used will vary with the particular cement and vzith where and how the agent is added to the slurry, but it is unlikely to be less than 10 ppm based on the dry solids content of the slurry. It is inadvisable to use more than 1000 ppm because of the danger of blocking the foraminous surface (sieve or felt). In practice, it has been found convenient to start by trying an addition of 600 ppm, and then experimenting with reduced quantities. We find that when adding the flocculant to a Hatschek type machine close to the de-watering stage, the quantity chosen should, generally be sufficient to give 50 to 600 ppm based on the dry solids content of the slurry.

It has been found important to ensure that the cement content of the slurry does not fall too low, because delamination of the product may then occur between the incremental layers built up on the machine. The flocculating agent enables sufficient cement to be retained to prevent delamination. It is also advisable to recirculate the filtrate or effluent from the machine, so as to reduce losses of fine materials and to maintain the cement content of the slurry at an adequate level, The glass fibre used in the reinforcement of the product, i.e. the chopped strands which retain their integrity, should be alkali resistant and have a long term durability in an alkaline cement matrix. One suitable material is that sold under the Trade Mark Chem-FIL by Fibreglass Ltd.

The single filament material should also preferably be alkali resistant but as it is primarily added to improve processing and not for long term reinforcement it can be a less -durable material if desired.

The proportion of glass fibre in chopped strand form in the cement composite produced on the machines will vary according to the properties desired in the finished product.

. At present costs quantities in excess of 10% by weight of 5 the finished product are unlikely to be competitive in the market place, and in any case no appreciable advantage is obtained by exceeding this proportion. We find in general that levels of the order of 5% by weight of the finished product give a product satisfactory for most purposes. It is unlikely that less than 2% will provide a satisfactory product for commercial production.

The proportion of single filament fibre to chopped strand reinforcing fibre will vary to some extent with the design of the machine being used and the characteristics of that machine. We find a proportion of 1:1 is a convenient ratio to use in the first place with any machine. In general any further guidance is difficult to give as the final proportion determined by further trial,if necessary on the plant, will also be affected by the rate at which the machine is operated as well as its design. It is obviously essential to take care that,e.g. in increasing the proportion of single filament fibre to improve the ease with which the product can be handled in the gresistate, care is taken not to adversely affect the strength properties of the finished product. The choice of the composition of the slurry for any particular machine must therefore be made following the general guide lines laid down above.

As stated above the mixes can be based upon ordinary Portland cement as the sole non-fibrous inorganic constituent and a portion of the cement can be replaced by fillers.

For example, limestone flour may be added both because 5 of its cheapness compared with ordinary Portland cement and also to reduce drying shrinkage. Lightweight aggregates such as vermiculite or perlite may he included to reduce the density of the product, with beneficial effects on its thermal insulation properties. Also, it is known that the addition of materials containing active silicon dioxide, sueh as pulverised fuel ash and kieselguhr may bring about a reaction with the lime which is liberated as the cement hydrates, and thus reduce the alkalinity of the product. This may be especially valuable because the reinforcing glass fibres may be subject to degradation by an alkaline environment.

Experiments were carried out, using the same method as that used in obtaining the data for Table 1, to compare the effects of such additives upon slurry filtration properties. The general batch formula was as follows! Percentage by weight Ordinary Portland Cement 68 Cement substitute 23 Chopped-strand alkali- resistant glass fibre 3 Glass wool, 7.5 Um diameter, λ shredded through 12 mm screen Polyacrylamide flocculating agent Added as 0.01% aque(Aquafloc 4103 ex Dearborn Chemicals Ltd) ous solution at 250 ppm of solids Water to give a water/solids ratio of 25:1 (including flocculant solution) Table 5 gives the results obtained when filtering the slurries through a 300//m mesh sieve, simulating the effects to be expected on a Hatschek asbestos-cement machine. 8 4 47 a 4) o 5 OT ri TABLE 5 % solids in filtrate transmitted through 300 pm υ a d < s § b β . * 3 o ¢4 § <; ti w « <2 υ ·* « b M rt rt « > ε «.j? Q b rt ΰ OS JO (2 *4 c rt E rt υ -rr b b ιΛ ca o CM • d co ca CM V u rt β b n rt (b b 0 V β tP 0 XT O £ X rt rH υ Ό c « b J3 0 » ΛΒ441? The effect of the additives on the percentage solids transmitted through the sieve did not vary greatly between the materials examined. Most served to reduce the sieve losses slightly, and only tele and pulverised fuel ash increased this factor.

There was a wider spread of de-watering times, however, with mica giving easily the fastest drainage.

All additives except ground blast furnace slag reduced the drainage time compared with pure cement.

All slurries formed complete webs on the sieve, though some (noticeably limestone) were of an uneven texture. The lightweight aggregates formed thick, spongy structures'as was anticipated. The large volume of perlite could not be wholly absorbed within the web structure.

It thus appears that these and similar materials may be used in conjunction with ordinary Portland cement to make glass fibre reinforced building boards on a Hatschek asbestos cement machine, using the method of the present invention, though adjustments to running conditions will be necessary to allot/ for the different drainage characteristics.

However, it is obviously essential that the additives used in partial substitution for cement should, possess sedimentation characteristics fairly similar to those of cement. For example, silica in the form of sand is not easily incorporated since the heavy particles quickly separate out from the mix and take no part in the formation of floes promoted by the polyacrylamide flocculant. Therefore attention must be paid to this factor when selecting materials for this purpose.

Claims (11)

1. A method of manufacturing an asbestos-free fibre-reinforced cement composite material in which a water-laid' web of cement and continuous-filament chopped glass fibre reinforcing 5 strands is deposited on a foraminous surface from a fluid 3lurry of those materials, and is de-watered by suction, wherein prior· to the web being laid a flocculating agent is added to the slurry, and in addition to the chopped strands there are present, in the slurry, single filaments io of inorganic non-cry3talline material. V

2. A method according to Claim 1, wherein the single filament material used is a glass wool or mineral wool.

3. A method according to Claim 1, wherein the single filament material used is produced from continuous filament 15 chopped glass fibre strands which separate or filamentise on contact with the cement slurry.

4. A method according to any one of the preceding Claims wherein the single filaments used have a diameter in the range from 0.5/«m to 20/zm. 2u 5. A method according to Claim 4, wherein the single filaments used have a mean diameter in the range from 4 to 9/ 6. A method according to any one of the preceding Claims, wherein the single filaments used are from 3 to 30 nun in 25 length. - 25 «8447 7. A method according to any one of the preceding Claims, wherein the single filament material is incorporated in an amount to constitute from 1% to 10% by weight of the total solids in the slurry.

5. 8. A method according to Claim 7, wherein the single filament material having diameters in the range 4 to 9ιλ m, is incorporated in an amount to constitute from 2% to 6% by weight of the total solids in the slurry.

6. 9- A method according to any one of the preceding Claims, 10 wherein the flocculating agent used is a polyacrylamide of anionic character.

7. 10. A method according to Claim 9, wherein the flocculating agent is added in an amount to constitute from 10 to 1000 parts per million of the dry solids content of 15 the slurry.

8. 11. A method according to Claim 10, wherein the flocculating agent is added in an amount to constitute from 50 to 600 parts per million of the dry solids content of the slurry, 20 '

9. 12, A method according to any one of Claims 9 to ll, wherein the flocculating agent is added to the slurry immediately before deposition of the web,

10. 13- A method according to any one of the preceding Claims, wherein the chopped glass fibre reinforcing strands are 25 incorporated in an amount to constitute from 2% to 10% of the dry solids content of the slurry.

11. 14. A method of manufacturing an asbestos-free fibrereinforced cement composite material according to Claim 1, substantially as hereinbefore described by way of example.