WO1990011976A1

WO1990011976A1 - Process for forming water resistant magnesian cement introduction

Info

Publication number: WO1990011976A1
Application number: PCT/AU1990/000132
Authority: WO
Inventors: John Ralston; Roger St. Clair Smart; Alexander Donald Mair; Julianna Csavas
Original assignee: Jeffcott Holdings Ltd (In Liquidation)
Current assignee: Jeffcott Holdings Ltd (In Liquidation)
Priority date: 1989-04-05
Filing date: 1990-04-05
Publication date: 1990-10-18
Anticipated expiration: 1991-10-05
Also published as: EP0469002A1; NZ233232A; EP0469002A4; CA2051414A1

Abstract

A process for the formation of magnesian cement having low water solubility and long term retention of mechanical properties. The process comprises acidulating insoluble basic, hydroxy and/or fluorophosphatic mineral material to activate it prior to or after mixing the mineral material with reactive magnesium oxide source material such as calcined dolomite or calcined magnesite, source material for magnesium chloride and/or sulphate such as bitterns, magnesium chloride hexahydrate or epsomite and sufficient water to make a workable slurry, with or without other common additives. The mineral material may be in the form of low grade phosphate rock and in some cases it may be desirable to calcine the phosphate mineral material prior to acidulation and mixing the cementitious components.

Description

PROCESS FOR FORMING WATER RESISTANT MAGNESIAN CEMENT INTRODUCTION

The present invention relates to a process for producing magnesian cementitious materials and particularly, but not essentially to the production of magnesium oxychloride hydrate and magnesium oxysulphate hydrate cementitious compositions often known as Sorel cements.

BACKGROUND OF THE INVENTION

1. Field of the Invention

Magnesium cements based on reaction of magnesium oxide with magnesium chloride or sulphate have been known for a long time. Such a cement formed with magnesium chloride, in particular, is commonly known as Sorel cement. This magnesium oxychloride hydrate cement, when cured, is generally characterized by the presence of the crystalline compounds 5Mg(0H)₂.MgCl₂.8H₂0 and/or 3Mg(OH)₂.MgCl₂.8H«0, the relative proportions of the two compounds depending on the stoichiometry of the cured mixture. Formation of the 5Mg(0H)₂.MgCl₂.8H₂0 compound is generally preferred. Often on aging, the cement, especially at the surface, carbonates to form another crystalline compound, Mg(0H)₂.MgCl₂.2MgC0₃.6H₂0. These magnesium oxychloride hydrate cementitious materials are renowned for their ultimate high strength and for the rapidity with which they attain such strength, but unfortunately suffer severe disadvantages which have discouraged wide use.

The foremost problem with magnesium oxychloride cements is their lack of water resistance, for, after

SUBSTITUTE SK£ET appreciable contact with water, drastic strength loss and even total disintegration can occur. Associated with the lack of water resistance is the problem of corrosion arising frojfn the large amounts of magnesium chloride leaching from the cement upon contact with water. The presence and liberation of soluble salts in the cured cements also causes efflorescence problems.

2. Description of the Prior Art

The prior art teaches that numerous attempts have been made to solve the problem of lack of water resistance in magnesium oxychloride cements. Many additives have been tested to improve the water resistance either by incorporation in the cementitious mix prior to setting or by application to the set or hardened cement. Such additives have included sodium silicates, polysilicates, ethyl silicate, waxes, silicones, stearates, linseed oil and phosphoric acid or its soluble salts. However, none of these additives have apparently led to commercially widespread acceptance of magnesium oxychloride cements except in a few selected areas such as speciality flooring.

Of the previously proposed additives incorporated in the magnesium oxychloride mix prior to setting phosphoric acid and sodium hexametaphosphate appear to be superior. R. Smith-Johannsen has recently concluded in U.S. Patent 4,352,694 that phosphoric acid is preferred over its soluble salts as an additive in obtaining a water resistant cured Sorel cement of sufficient strength for use in the building industry. However, phosphoric acid is relatively expensive and its addition to the magnesian oxychloride cement may make the cement commercially unviable. Addition of sparingly soluble or substantially insoluble phosphate has also been proposed for improving the properties of Sorel cements. These include phosphates or secondary phosphates of calcium, magnesium and other alkaline earth metals, zinc, aluminium and copper ⁽US Patents 2,351,641, 4,185,066 and 4,158,570). However, all of the described phosphates are either aci_d or neutral phosphates. Basic phosphates or hydroxyphosphates, such as crandallite, millisite, wavellite and hydroxyapatite, and fluorophosphates, suc_h as fluorapatite, all found occuring naturally as phosphatic minerals, have not been proposed for improving the properties of Sorel cements due to their high insolubility and inertness which results in no benefit whatsoever in improving the properties of Sorel cement.

SUMMARY OF THE INVENTION

^{It is} an object of the present invention to provide an improved process for producing a water resistant magnesian cementitious material.

According to the present invention, a water resistant magnesian cementitious product is pro_duce_{d b}y admixing particulate phosphatic material with a reactive magnesium oxide source material, a magnesium chlori_de and/or magnesium sulphate source material and sufficient water to provide a workable slurry and setting the slurry, and wherein the phosphatic material comprises an insoluble basic, hydroxy and/or fluoro phosphatic minera_l material which is activated at least by partial acidulation.

Further according to the present invention there is provided a magnesian cementitious product when produce_d by the process described in the immediately preceding

SUBSTITUTE SHEET paragraph.

By the present invention insoluble phosphatic mineral material can be readily incorporated into a magnesian cementitious material to promote water resistance and superior long term mechanical properties. This has the advantage of being able to utilize relatively cheap phosphatic source material to improve the physical properties of magnesian cements.

As foxnd in the present invention, addition of the insoluble basic, hydroxyphosphate and/or fluorophosphate mineral materials to the cement mix confers no water resistance benefit to the cured cement unless they can be activated. The water solubility of a magnesian cement incorporating such inactivated insoluble mineral materials is in the range 25 to 30%, usually about 27%. Solubility is a measure of percentage weight loss after being subjected to a water treatment. In this invention it has been discovered that such beneficial activation of insoluble phosphatic minerals, particularly basic, hydroxy- and/or fluoro-phosphates although others may be appropriate, can be achieved by partial acidulation, optionally with a pre-calcining step. The partial acidulation of the phosphatic mineral material may be carried out before, during or after its addition to the cement. After such activation, the presence of the phosphatic material in the cement confers a significantly increased water resistance to the cement of 20% or less, preferably about 10% or less and most preferably about 7% or less. A magnesian cement having a water solubility of the order of 10-20% may be most appropriate as a filler but at lower levels may be readily used in structural situations. In addition to the advantageously low water solubility characteristics, the mechanical properties, particularly compression strength, may only degrade slightly over extended periods of time, in some cases only _by a maximum in the order of 20 to 25% and commonly 10 to 15% or less.

The admixing may generally be performed at ambient or room temperature, but there may be circumstances where it is advantageous to mix at elevated temperatures, and by way of example only attention is directed to the process and product described in International Patent _Application WO 87/04145, the content of which is incorporated herein by reference, which process and product may be modified in accordance with the present invention.

_Advantageously, the magnesian cement is produced _from _low cost or byproduct magnesium compositions, such as bitterns as a source of magnesium chloride and low gra_de magnesite as a source of magnesium oxide. Bitterns is the residual liquor from the controlled evaporation of seawater. It may not be necessary to provide additional water in the mixture to give the slurry a workable consistency if the source materials comprise sufficient water. If the magnesium chloride source material is particulate it is preferably finely divided, with, for example, a particle size of less than about 250 micron.

It has been discovered that the use of a low sodium content _bitterns with a Mg/Na weight ratio of greater than three, preferably greater than eight, is particularly advantageous in that undesirable efflorescence in the resultant magnesian cement may be markedly reduced.

With regard to the magnesium oxide source material, a magnesium carbonate mineral such as magnesite or dolomite can be utilised after calcination at sufficiently low temperatures to provide sufficient reactivity in the resultant MgO. Magnesite is preferably calcined at temperatures of less than about 820°C to retain sufficient surface area and reactivity in the product MgO. Dolomite is preferably part calcined to form MgO and calcite before use. Those skilled in the art will appreciate that other sources of MgO such as calcined brucite or calcined magnesium hydroxide, as can be derived from seawater, or mixtures of any of the above, are also suitable for the practice of the process of the present invention. Preferably the magnesium oxide source material is finely divided.

A phosphatic material comprising as a major portion calcian hydroxyphosphates and/or fluorophosphates may be used in the preferred embodiment. Such a composition is found ocσuring naturally in commercially mined phosphate deposits substantially as the mineral apatite, either fluorapatite Ca₅(P0.)₃F, hydroxyapatite Ca₅(P0₄)₃0H, or as a mixed fluoro-hydroxy derivative. In sedimentary deposits, appreciable substitution of the phosphate by carbonate often occurs in the apatite crystal structure, to give carbonate apatites (francolites); these also are suitable compositions, as are chloro apatites.

Another composition which may be used in the preferred embodiment is the waste phosphatic material found occuring naturally as the so-called "leached-zone" phosphates in weathered sedimentary phosphate deposits at locations including Florida (USA), Senegal (West Africa) and Christmas Island (Indian Ocean). This phosphate, often a mixture of apatite and hydroxyphosphate minerals such as crandallite and millisite, is a low grade ore which cannot at. present be economically converted to other useful products.

Preacidulation of the insoluble phosphatic material, either¹ partially or wholly, is advantageously performed using phosphoric acid, sulphuric acid or a combination of the two. Use of phosphoric acid to activate the insoluble phosphatic material can beneficially provide water resistance and mechanical strength superior or substantially equivalent to that achieved using a greater amount of phosphoric acid alone, without the presence of the insoluble phosphatic material. Use of sulphuric acid to preacidulate the insoluble phosphatic material provides water resistance and mechanical properties essentially equivalent to that obtained using phosphoric acid, beneficially allowing total replacement of expensive phosphoric acid by much cheaper sulphuric acid.

Alternatively, the phosphatic material in the mixture may not be preacidulated as specified in the immediately preceding paragraph but acidulated at a later stage when the phosphatic component is already bound within the set cementitious matrix, whereby the cement is immersed briefly in an acidic solution to at least partially acidulate that phosphatic mineral component within, or at least close to the surface of, the cementitious object. Examples of appropriate acid solutions are phosphoric acid and sulphuric acid.

In another embodiment of the present invention, the pre- or post- acidulation of the insoluble phosphate mineral component to provide a cementitious product of increased water resistance is carried out after mildly precalcining the insoluble phosphatic component prior to mixing with the other cement components. Such precalcining may be performed at temperatures in the range of, for example, 300 to 700°C, preferably 450 to 550°C for upto about 3 hours. Where the insoluble phosphatic component comprises, for example, crystalline calcium aluminium hydroxyphosphates such as crandallite and millisite as in the aforementioned low grade ores, calcination for instance at 500°C with a retention time of one hour has been found sufficient to destroy the

TIT T ET mineral crystallinity, converting the mineral phases to a more reactive amorphous structure. This precalcining step benefits the use of acids such as hydrochloric and citric in the acidulation step, acids in which calcium aluminium hydroxyphosphate minerals such as crandallite and millisite are relatively insoluble unless mildly calcined to destroy their crystallinity.

In activating the phosphatic mineral by acidulation, it is preferred to only partially acidulate, especially when significant aluminous phosphate is present. As the optimum degree of acidulation depends in part on the absolute amount and composition of the phosphate incorporated or to be incorporated in the cement, the nature of the acid used and the degree of phosphate precalcination, if any, it must be determined by trial and experiment. Complete acidulation and solubilization of the phosphates often confers no additional benefit and may in fact be detrimental to cement properties, including that of water resistance. Acidulation times will commonly be from 1 to 60 minutes, preferably from 4 to 20 minutes, most preferably about 15 minutes. The amount of acid appropriate for the acidulation is dependent upon the amount of reactive MgO in the mixture and will generally be upto about 20g per lOOg MgO.

Table 1 represents advantageous operating parameters and variables which are presently considered appropriate for the magnesian cement production process using Christmas Island mineral phosphatic material preacidulated with phosphoric acid or sulphuric acid and with optional precalcination of the phosphatic material. TABLE 1

Process Conditions and Operating Preferred

Parameters Range Range

Preacidulation with Mineral Phosphatic Component

P Δ0o,- content

(weight percent) 10-38 25-38 Feed size (micron) <250 <125

Preacidulation acid (g/lOOg MgO):

Phosphoric acid 1-20 3-6

Sulphuric acid (70%) 1-10 4-8 Preacidulation duration

(minutes) 1-60 4-20

Calcination conditions (where applicable)

Calcination temperature (°C) 300-850 400-600 Calcination duration (minutes) 0.1-180 15-60

Cementitious mixture

MgO/MgCl₂ stoichiometry, moles 4-12 6-8

Phosphatic mineral component (g/lOOg MgO) 2-100 5-25

Feed size, MgO (micron) <250 <90

2 Surface area, MgO (m /g) 5-80 25-60

Mg/Na weight ratio, bitterns >3 >8

(where applicable)

By operating within these parameters, the stoichiometry of the process is controlled to provide as a major crystalline cementitious phase formed in the hardened cement the magnesium oxychloride hydrate

SUBSTITUTE SHEE^" compound 5Mg( 0H ) ₂.MgCl₂. 8H₂0.

In all embodiments, additives and fillers may be incorporated into the cementitious mix prior to setting to increase.water resistance and/or mechanical strength, to provide suitable colouring or texture, to control shrinkage or expansion, to control the rheological properties of the cementitious slurry, or simply to act as inert extenders. Such additives which might be used include, but are not limited to, inorganic and organic fibres, pigments, lignins, lignosulphonates, surfactants and foam promoters, superplasticizers, sawdust, woodchips, bagasse, rice hulls, slags, flyash, talc, silica sand, clays and other aluminosiliσates. It has been found of particular value to incorporate silica fume or microsilica in the cementitious mix to further promote water resistance in the hardened cement. In all embodiments, the magnesium chloride source material may be replaced wholly or in part by a magnesium sulphate source material such as epsomite.

The magnesian cementitious composition of the present invention may have wide applicability in the building and other industries, and may beneficially be incorporated in many products including foamed insulation panels •and coatings and as a binder in particle- and hard-board compositions.

EXAMPLES

The following examples are given by way of illustration and not by way of limitation since various changes therein may be made by those skilled in the art without departing from the true spirit and scope of the preseηt invention. Unless stated otherwise, Christmas Island phosphate rock comprising a mixture of calcium aluminium hydroxyphosphate minerals and apatite was used - 11 - in the Examp^les, ground to 125 microns or less. _The use of magnesium chloride hexahydrate simulate_d the incorporation of bitterns in the process.

EXAMPLE 1

A cementitious slurry was prepared by mixing 125 parts of low grade calcined magnesite, containing 100 parts of MgO, with 78 parts of magnesium ch_lori_de ^hexa^hy^drate plus 35 parts of added water. T_he magnes_ite used, from Copley, South Australia, was finely ground a^fter calcination at about 810^βC to pass 90 microns and possessed a surface area of about 50 m²/g after calcination.

In a series of statistically designed experiments usⁱng such preparations, it was found t_hat addition to the alkaline cementitious slurry of 8.3 parts calcine_d Christmas Island C-grade phosphate rock (31 wt. % p o ₎ was significant ⁽at the 96 % confidence level₎ at ^ ^decreasing the water solubility of the resultant hardene_d cement to 25.7 % when compared to control preparations containing no such Christmas Island phosphate rock, t_hese contro^ls ex^hibⁱting an absolute water solubility of 27.0 % as har^dened cements. In contrast, a_ddition of t_he Christmas Island phosphate when in an uncalcined state provided no enhancement to the water resistance whatsoever. The solubility test comprised immersing sma^ll samples of the hardened cements in tap water ⁽Adelaide, South Australia) for 17 hours at 80°C, this treatment satisfactorily simulating long term water immersⁱon at ambient temperatures. The ca_lcination treatment of the Christmas Island phosphate comprise_d heating the rock in a muffle furnace for one hour at 500-C, whereby the crystallinity of the major aluminous phosphate minerals in the rock, namely crandallite and millisite, is destroyed, converting the a_luminous

SUBSTITUTE SHEET - 12 - phosphate to more reactive amorphous _forms.

EXAMPLE 2

In a first cement preparation, an acidulated s_lurry was prepared from 16.7 parts of Christmas Island C-grade phosphate rock (25.2 wt. % P^) and 4.2 parts of phosphoric acid. Water was added (3 parts) to maintain a satis^factory consistency and prevent the mixture _from ^becomⁱng too thick. The acidulated mixture was _le_ft _for 15 minutes wⁱth occasional stirring. Ca_lcine_{d C}op_ley magnesite ⁽125 parts containing 100 parts MgO), magnesium chloride hexahydrate (78 parts) and further water (32 parts⁾ were next added to the slurry. _After stirring, the cementitious slurry was poured into molds and allowed to set.

^Two ^furt^her preparations were made in a similar manner without addition of Christmas Islan_d phosphate by mixing 125 parts of calcined Copley magnesite, 78 parts o^f magnesium chloride hexahydrate and 35 parts o_f water. To one of these further preparations, 7 parts of phosphoric acid was additionally incorporated.

^After allowing to harden for at least 15 days, the cements from all three preparations, in the form of 2 cm cubes, were subjected to compressive strength testing and ^determination of water resistance by the solubility test at 80°C specified in Example 1.

^A sample from that preparation containing no Christmas Island phosphate and no phosphoric acid showed a compressive strength of 50 MPa with no water treatment, but water treatment at 80°C caused an identical sample from the preparation to experience almost total disintegration, with an accompanying weight loss of 30 %. A sample from the preparation containing no Christmas Island phosphate but containing 7 parts of phosphoric acid showed a compressive strength of 37 MPa with no prior water treatment, while a sample from the same preparation after the 80°C water treatment exhibited a compressive strength of 23 MPa and a water solubility of 5.5 %.

For samples from the preparation containing both Christmas Island phosphate and phosphoric acid, a compressive strength of 38 MPa was obtained with no prior water treatment and a value of 30 MPa with an associated water solubility of 5.0 % for an identical sample subjected to the 80°C water treatment.

This example demonstrates that presence of phosphoric acid, when contrasted to its absence, provides high strength retention and low solubility for hardened cements after a severe water treatment. This example unexpectedly demonstrates further that presence of insoluble phosphatic minerals after partial preacidulation with phosphoric acid is beneficial in that it can provide greater or equivalent strength retention after water treatment than does use of a greater amount of phosphoric acid alone. By contrasting with Example 1, the present example also demonstrates by the comparative water solubilities of the cements that use of the preacidulation step for the insoluble phosphate is superior to use of insoluble phosphate without preacidulation.

EXAMPLE 3

An acidulated slurry was prepared from 16.7 parts of Christmas Island C-grade phosphate rock (25.2 wt. % P.O,.) and 6.3 parts of 70 % sulphuric acid. Sufficient water was added (6 parts) to prevent the mixture from becoming

SUBSTITUTE SHEET too thick. The acidulated mixture was left for 15 minutes with occasional stirring. Calcined Copley magnesite (125 parts containing 100 parts MgO), magnesium chloride hexahydrate (78 parts) and further water (36 parts) were next added to the slurry. After stirring, the cementitious slurry was poured into molds and allowed to set. Tests performed on the hardened cements were identical to those specified for Example 2.

A compressive strength in excess of 54 MPa was exhibited by a 2 cm cube of this cement with no water treatment, while a water treated sample (as before at 80°C) possessed a compressive strength of 21 MPa after associated dissolution of the sample during the water treatment of 7.3 %.

This example illustrates a further aspect of the preferred embodiment of the present invention. It unexpectedly demonstrates that despite the higher water solubility (compared to those cement samples in Example 2 containing phosphoric acid) probably due to leaching of the calcium sulphate formed, a high strength is retained after water* treatment, comparable to that water treated sample in Example 2 wherein phosphoric acid was used alone without mineral phosphate addition. The present example further demonstrates that for equivalent water resistance as quantified by compressive strength retention in the cured cement, sulphuric acid, when used in conjunction with an insoluble phosphate mineral component, can beneficially replace the more expensive phosphoric acid used alone.

EXAMPLE 4

Three series of cements were prepared in this example. All cement compositions were identical except for the phosphate component, and a slightly varying water content to provide a suitable consistency _during mixing. In one series, differing amounts of phosphoric acid were adde^d. In the other two series a constant amount of Christmas Island C grade phosphate rock was added toget^her with differing amounts of phosphoric acid ₍the second series⁾ and sulphuric acid (third series). All cements also contained silica fume. The 100 parts of MgO used in each formulation was contained in 125 parts calcined Copley magnesite.

In the first series a premix was prepared by addition of 29.4 parts MgO to 30.6 parts water followed ^by selected additions of phosphoric acid (as given in Table 2⁾. The premix was left to stand for 15 minutes a^fter whⁱch 69 parts MgCl₂.6H₂0 was added _followed by the final 70.6 parts MgO premixed with 3.7 parts silica fume.

In the second and third series, selected additions of phosphoric acid and sulphuric acid, respectively, as detailed in Table 2, were added to 14.7 parts uncalcine_d phosphate rock followed by a small amount of water if the mixture was dry. The premix was left to stan_{d f}or 15 minutes after which 38 parts water was added, followe_{d b}y 29.4 parts MgO. This was left to stand for 5 minutes. ^At this point 69 parts MgCl₂.6H₂0 was adde_d, _followed _by 3.7 parts silica fume premixed with 70.6 parts MgO.

Water solubilities were determined at 80°C by the method given in Example 1. Results are shown in Table 2.

The data in Table 2 demonstrates that the water solubility for cements containing acidu_late_d roc_k phosphate goes through a minimum with an optimum amount o^f acid ⁽for both phosphoric and sulphuric acids₎ at about 5.5 parts with the compositions used in this

Example. Higher amounts of acid under these conditions are detrimental, decreasing the water resistance.

SUBSTITUTE SHEET Solubilities with sulphuric acid are higher than those with phosphoric acid, in part, at least, due to the formation of appreciably soluble calcium sulphate with the forme .

It is further beneficially demonstrated (as in Example 2) that the degree of water resistance imparted by phosphoric acid alone can be matched by a significantly lower amount of phosphoric acid if used in combination with rock phosphate. For example, water resistance achieved by 7.4 parts phosphoric acid is equalled by using only about 2.8 parts of phosphoric acid in combination with 14.7 parts rock phosphate addition.

Silica fume enhances the water resistance of the cements. A 22% weight loss found with silica fume addition but with no phosphate is lower than the 28% found in control cements. The beneficial effect of silica fume is maintained in phosphate containing cements, with the water resistance contributions apparently substantially additive as seen by comparing the water solubility data of this Example with Examples 2 and 3.

TABLE 2

Amount of Acid Added and Correspondinσ Solubilitv

Solubility (weight loss, percent)

Amount of Acid Series: JL 2 3

(parts/100 pt MgO) Phosphoric Rock Phosphate Rock phosphate

Acid Alone & phos acid + sulphuric

0 21.2 22.8 22.0

0.9 8.5 5.9 10.6

1.8 5.2 3.0 8.6

2.8 5.2 2.6 7.0

3.7 3.5 1.9 4.5

5.5 2.7 1.6 4.0

7.4 2.2 1.7 4.7

TIT TE Sr EE^" EXAMPLE 5

Two types of cements were prepared from the high grade calcined magnesite from Kunwarara, Queensland: a water resistant cement containing sand, rock phosphate and phosphoric acid and a control cement containing no additive other than sand.

For the phosphate modified cement, a dry mix was prepared from 1170 g of finely ground Kunwarara calcined magnesite (calcined 800°C; free MgO 93.5 wt. %) and 5 kg of dry sharp quality building sand in a mixer of commercial design. A premix was separately prepared by adding 50 g phosphoric acid to a slurry of 200 g

Christmas Island C grade phosphate rock and 100 ml water. After 15 minutes, 130 g of Kunwarara calcined magnesite was added to the phosphate premix with stirring. This mixture formed a dry mix and was subsequently slurried with another 100 ml of water, and washed into the MgO- sand dry mix with 1000 ml of magnesium chloride solution (density 1.30, 32 % MgCl₂). Minimal additional water was added where necessary to maintain satisfactory consistency in the cement slurry. The cement slurry was poured into 10 cm cube molds, with vibrating to reduce porosity.

A control cement was similarly prepared by making a dry mix of 1300 g Kunwarara calcined magnesite and 5 kg of dry sand. To this mix was added 1000 ml of the same magnesium chloride solution. Additional water was added to the slurry as required.

The phosphate modified cement exhibited compressive strengths in the dry state of about 59 ± 2 MPa.

Corresponding compressive strengths after immersion in tap water at room temperature for 7 and 30 days were still high, at 42 and 49 MPa respectively. _A wet/_dry cyclic test at ambient temperature, consisting of 8 cycles of one day in fresh tap water followed by two days stored in air, resulted in a compressive strength in a 10 cm cube of 57 MPa, practically unchanged from that for continued storage in air. The relatively small change in compressive reflects the minimal solubility of the samples noted during the tests.

For cubes of the control cement containing no phosphate, typical compressive strengths for dry storage reached 90 MPa. However for identical water immersion and cyclic testing done for the phosphate modifie_d cement, the control cubes disintegrated; no compressive strength testing was possible.

This example demonstrates a novel method o_f incorporating the acidulated phosphate in the cement, whereby the phosphate slurry is converted to a fine dry powder by premixing the slurry with a portion of t_he Mg_O used in the cement. This dry premix is a convenient way of storing the acidulated phosphate and allpws it to _be added to the cement mix at a convenient later time. The dry premix could also be added back with the remainder of the MgO, for later mixing with the magnesium chloride.

EXAMPLE 6

A series of cements were prepared with Christmas Island Grade C phosphate rock from which all the apatitic phosphate component had been removed. The apatite was removed by repeating leaching of the rock with 2.5 molar hydrochloric acid solution, while monitoring the _leach process products by X-ray diffraction analysis. The X- ray diffraction analysis showed that all apatite originally present had been removed and that the major crystalline phases present in the leached rock were the ca^lcⁱum a^lumⁱnium hydroxyphosphates, crandallite an_d millisite; these had remained undissolved in the leaching step.

Using a ^factorially designed series of cements, hal_f of the samples prepared incorporated this leached phosphate in the precalcined state; the remaining samples ⁱncorporated the leached phosphate in an uncalcined state. The calcining step used a one hour t_hermal treatment at 550°C in a muffle furnace. Just prior to incorporating in the cement, the extracted phosphate roc_k was acidulated with either concentrated hydrochloric or 70% su^lphuric acids for periods ranging from 5 to 10 minutes. The acidulated phosphate was then mixe_d wit_h 125 parts o^f finely ground calcined Copley magnesite ⁽containing 100 parts free MgO), 63 parts of Mg_Cl₂.6H₂₀ and a^bout 42 parts water. The amounts of extracted phosphate rock and of acid used was systematica_lly varied to ^be 12.5 or 25 parts/100 parts free MgO and 4.2 or 12.5 parts/100 parts free MgO, respectively.

Water solubility tests were made at 80°C (as detailed in Example 1) on the cements after one week maturation and again another six weeks later. _No sign^ificant differences were observed in t_he two sets of solubility measurements made at different times.

The solubility data showed that water solubility was low if calcined leached phosphate rock was used in the cement, averaging 4.6% and 4.5% weight loss for hydrochloric and sulphuric acid, respectively, as ac^idu^lating agent. However, when using uncalcined leached phosphate rock, the corresponding average weig_ht losses were much higher, at 18.7 and 13.4%, respectively. This compares to the 25-30% weight loss experience_d in control cements with no phosphate addition.

SUBSTITUTE SHEET These results demonstrate that if the readily acid soluble apatite (calcium phosphate) impurity is removed, the calcium aluminium hydroxyphosphate minerals remaining must be activated by calcination prior to acidulation to obtain minimal water solubility and maximum water resistance in the cement. Calcination converts the crystalline hydroxyphosphate to a more acid soluble amorphous state. Considerably less water resistance is imparted to the cement by the calcium aluminium hydroxyphosphate component if uncalcined; the results demonstrate that hydrochloric acid is less effective with the uncalcined phosphate in this instance than is sulphuric acid.

Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is to be understood that the invention includes all such variations and modifications which fall within the its spirit and scope. The invention also includes all of the steps, features, compositions and compounds referred to or indicated in this specification, individually or collectively, and any and all combinations of any two or more of said steps or features.

^"

Claims

CLAIMS :

1. A process for forming a water resistant magnesian cementitious product which comprises admixing particulate phosphatic material with a reactive magnesium oxi_de source material, a magnesium salt source material, the salt being selected from one or both of c_hloride an_d sulphate, and sufficient water to provide a workable slurry and setting the slurry, and wherein t_he p_hosp_hatic material comprises an insoluble basic, hydroxy an_d/or ^fluoro phosphatic mineral material which is activated at least by partial acidulation.

2. A process according to Claim 1 wherein aci_dulation is performed on the phosphatic mineral material using phosphoric acid, sulphuric acid or a combination of the two acids.

3. A process according to Claim 1 wherein acidulation of the phosphatic mineral material is performed prior to mⁱxing with the reactive magnesium oxide source material and the magnesium salt source material.

4. A process according to Claim 3 wherein a substantially dry premix comprising the acidulated phosphatic mineral material with some or a_ll o_f t_he reactive magnesium oxide source material is mixed with the magnesium salt source material and sufficient water to provide the workable slurry.

5. A process according to Claim 1 wherein the phosphatic mineral material is calcined prior to acidulation and to mixing with the reactive magnesium oxide source material and the magnesium salt source material.

6. A process according to Claim 1 wherein the

SUBSTITUTE SHEE^' phosphatic mineral material comprises calcian hydroxyphosphates, calcian fluorophosphates or mixtures of the two.

7. A process according to Claim 6 wherein the phosphatic mineral material comprises calcium aluminium hydroxyphosphates.

8. A process according to Claim 1 wherein the magnesium salt comprises magnesium chloride which is present as magnesium chloride hexahydrate.

9. A process according to Claim 1 wherein the magnesium salt comprises magnesium chloride which is present as bitterns.

10. A process according to Claim 9 wherein the bitterns has an Mg/Na ratio greater than 8.

11. A process according to Claim 1 wherein the reactive magnesium oxide source material comprises a calcined magnesium carbonate mineral selected from one or both of magnesite and dolomite, calcined brucite, calcined magnesium hydroxide or mixtures of any two or more.

12. A water resistant magnesian cementitious product comprising a cured reaction product of a slurry of a reactive magnesium oxide source material and a magnesium salt source material with water, said salt being selected from one or both of chloride and sulphate, said cured reaction product incorporating phosphate in its lattice structure which is derived from at least partially acidulated basic, hydroxy and/or fluorophosphatic mineral material.

S B TITUTE HEET