摘要:

L. A. THOMAS N. WOOSTER AND W. A. WOOSTER 345 PROBLEMS OF CRYSTAL GROWTH IN BUILDING MATERIALS BY F. M. LEA AND R. W. NURSE Received 1st February 1949 The growth of crystals is of interest in various directions in the study of building materials and structures. The present paper will be limited to problems arising in the manufacture of building materials and to some crystalline changes which arise subsequently when the material is in use. Crystallization at High Temperatures.-Industrial materials such as slags and cements are produced by sintering or fusing the raw materials at high temperatures and there is considerable technological interest in the extent and manner of crystal growth. With slags and high alumina cement the mix becomes completely molten and crystallization takes place from the liquid phase but with Portland cement only some 20-30 yo of the mix becomes liquid and crystal growth occurs both as a result of solid- liquid reactions and of direct crystallization from the liquid.The study of the phase equilibria diagrams is an essential feature of work on crystal growth in such materials for it enables the order of appearance on cooling of different crystals to be defined and the effect of departures from equilibrium to be traced. The complicated crystallization paths which can arise in polycomponent silicate systems have been discussed by various authors.1 2 They frequently involve disappearing phases and produce typical crystalline structures such as eutectic and peritectic patterns corrodcd crystals and zoning etc.Attention has also been drawn to the possibility of " independent crystallization " of the liquid phase taking place when the cooling process is rapid or involves two or more distinct stage^.^ The final structure in such cases of frozen equilibrium can be predicted from the phase equilibrium diagrams. Crystallization from the Melt.-A most noticeable effect in pofy- component silicate and aluminate systems is the tendency of particular compounds when crystallizing from the melt to appear as spheres (Fig. I). This effect is not related to the symmetry being very obvious in the case of zCaO.SiO, which has quite low symmetry. Optically the crystals Bowen The Evolzttion of the Igneous Rocks (Princeton 1928). 2 Hall and Insley Phase Diagrams for Ceramicists (Amer.Ceram. SOC. 1947). 8 Lea and Parker Phil. Trans. 1934 234 I. Lea and Parker Bzrilding Research Tech. Paper No. 16 (H.M. Stationery Office 1935). Parker and Nurse .J. SOC. Chem. I n d . 1939 58 255. M* FIG. 1.-Periclase (MgO) crystals in glass. ( x 250) To face page 3451 CRYSTAL GROWTH I N BUILDING MATERIALS 346 appear to be single individuals and it is only at a later stage when the crystals are larger or under conditions of slower growth that normal faces and forms appear. I t is generally considered 6 that curved or vicinal faces arise when steep concentration gradients exist in the solution or when the rate of diffusion is low. This can result from a high viscosity of the melt or from a high rate of crystallization.Spherical growth does not occur in melts of high silica content such that glasses are readily formed but it is charac- teristic of compounds such as CaO MgO zCaO.Si0 and 3Ca0.Mg0.zSi02 and to a lesser extent 3Ca0.A12.0 and 5Ca0.3A1203 which either do not form glasses or are easily devitrified. Spinel which readily foms a glass always develops as minute octahedra (Fig. 2 ) . If instability of the glassy state is to be associated with a high crystallization rate it would appear that the latter as well as the viscosity of the melt is an important factor determining the growth of spherical crystals. Growth of crystals by a solid-liquid reaction does not lead to spherical forms. Thus 3CaO.Si0 formed in this way shows its true symmetry but small crystals grown from a suitable melt are often spherical.Minerals differ much in their rate of crystallization and this frequently leads to complication of the crystallization path and to changes in the form of crystal growth. Where an incongruently melting compound occurs in a system the phase first crystallizing may react with the liquid at a definite temperature to form a new species. Frequently the new crystal grows at such a rate that the dissolving primary crystals are enveloped and excluded from contact with liquid so that on complete solidification the second phase may still contain inclusions of the first. Typical cases are inclusions of CaO in 3Ca0.A1,03 (Fig. 3) and MgO in monticellite. If such a formation is held for a sufficient length of time at a temperature just below the saturation temperature of the second phase the inclusions disappear.The rate of diffusion in solid phases at such temperatures must therefore be quite high. Growth of Single Crystals.-Work on the X-ray structure of cement minerals has been hindered in the past because the low symmetry neces- sitates single crystal determinations and methods for growing the crystals had not been devised. Many of the minerals melt incongruently or as in the case of 3CaO.Si0 decompose below the melting point and cannot therefore be grown from melts of their own composition by conventional methods. Attempts to encourage crystallization by the addition of mineral- izers have sometimes been successful but many failed because not enough was known of the crystallization paths in complex systems.Le Chatelier for instance was unable to crystallize 3CaO.Si0 from melt containing CaC1,. This subject has been taken up again by one of the present authors and single crystals of pure 3CaO.Si0 up to I mm. and 3CaO.Si0 solid solution up to 5 mm. in length have been grown. Twinned crystals of p zCaO.Si0 up to I cm. long have also been obtained (Fig. 4 and 5). The method used for 3CaO.Si0 is briefly as follows. Fig. 6 is a schematic diagram of the system CaO - zCaO.Si0 - CaCl,. A composition such as denoted by B (35 % 3Ca0.Si02 35.% y zCaO.SiO, 30 yo CaCl,) was made up from the previously reacted sllicates. A total weight of 5 g. of the mixture was heated in the electric muffle at a temperature of 15ooOC in a platinum crucible 2.5 cm.diam. and 2-5 cm. tall. After a period of about 4 hr. most of the chloride had evaporated; the melt was then cooled in air and " dusted " owing to the @? inversion of aCaO.Si0,. 6 Wells Ann. Re;borts 1946 84. Nurse 21st Cong. Ind. Chew. (Brussels 1948). FIG. 2.-Spinel (MgO.AI,O,) crystals in glass. ( x 250) FIG. 3.-Inclusions of CaO in 3Ca0.Al,0E. ( x 250) [To fat page 346 FIG. 4.-Single crystals of gCaO.Si0,. ( x 20) FIG. 5.-Twinned crystals of p aCaO.Si0,. ( x 20) F. M. LEA AND R. W. NURSE 347 The large crystals of 3CaO.Si0 were then separated by washing with alcohol on a 300-mesh sieve. Neglecting the decomposition and oxidation of CaCl into CaO and Cl, the melt composition follows the line BA in Fig. 6 and eventually passes into the 3CaO.Si0 primary phase field.Crystals of 3CaO.50 begin to form when the 1500’ C isotherm is reached and since the melt is then losing both CaC1 and 3Ca0.Si02 the melt composition follows the isotherm towards the aCaO.Si0 field. When all the chloride has evaporated the melt has composition A and consists of large crystals of 3CaO.Si0 and zCaO.Si0,. Similar methods have been used to prepare single crystals of a new compound 3SrO.Si0 from the oxides and SrCl,. It is not isomorphous with 3Ca0.Si02.* The successful preparation of spinel boules in the Verneuil furnace suggests that it should be possible to grow single crystals of congruently melting compounds such as CaO.A1,0, rzCa0.7Al,03 ~Ca0.A1,03.Fe,03 by [this method and it is hoped to attempt this shortly.CaO ( 1 5 7 0 ~ ) for the sake of clarity the older designation will be used. CaCL2 (7800~) FIG. 6.-System CaO - zC.aO.SiO?- CaC1,. Crystallization from the Glass .-Most studies of crystallization from the glassy state lo have been carried out on ccmmercial glass compositions. In such cases it has been assumed and scmetimes confirmed experimentally that the phase crystallizing is that to be expected from the relevant phase equilibrium diagram. A number of observations for which there is as yet no connected theory indicates that this is not always the case. In the crystallization of high alumina cement melts a large amount of a phase which is known as ‘ I unstable 5Ca0.3Al2O ” * (Fig. 7) often appears. Rankin and Wright found that both 5Ca0.3A120 and 3CaO.5A1,O3 occurred * Nurse unpublished data.Morey Properties of Glass (Rheinhold Pub. Corp. 1938). l o Morey Trans. Faraduy Soc. 1941 37 209. * There seems to be little doubt that the stable phase identified by Rankin and Wright in the system CaO - A1,0 is reCa0.7Al,03 and not 5Ca0.3A1,03. However CRYSTAL GROWTH I N BUILDING MATERIALS 348 IOOOO in unstable forms under certain conditions of cooling during their studies in the system CaO -A1,0,. Sundius l1 separated a mineral from high alumina cement clinker which corresponded closely in optical properties to Rankin and Wright’s unstable 5Ca0.3A1,03; the separation was not complete and his analysis showed that the mineral might be 3Ca0.2AI,03. Recent experiments at the Building Research Station indicate a composition higher in lime as being more likely but still not conforming in composition to any compound known in the system CaO - A1,0,.Dyckerhof l2 obtained “ unstable 5Ca0.A1,03 ” by annealing glass of that composition at about C. These experiments have been repeated a t the Building Research Station but owing to the dendritic nature of the crystals (Fig. 8) it was not possible to determine whether any glass remained. It was found however that the “unstable” compound was formed only below 1020’. The crystals in high alumina cement clinker are so well developed that it is difficult to believe that they are derived from devitrification of a glass ; they frequently occur in association with the stable form.A possible explanation is that a low-temperature stable form of 5CaO.- 3A1,03 exists or that at low temperatures 5Ca0.3A1,O3 decomposes into CaO.AlqO and 2Ca0.A120, but there is little evidence to support this since crystalline 5Ca0.3Al20 cannot be converted to the unstable form by annealing at any temperature. Furthermore the new species should crystallize from the melt in any polycomponent system in which the liquidus tem- perature fell below the decomposition or inversion temperature of the normal form of 5Ca0.3AI2O,. Liquid temperatures in the system 5Ca0.3A120 - Na,O.WO fall below IOZO’ but the primary 5Ca0.3Al20 still occurs in the stable form.8 There is some solid solution however so there is a possibility that the inversion (decomposition) temperature has been lowered.If these aluminates prove to be truly metastable phases this phenomenon may be of some interest in connection with recent studies of glass structures. Lukeshl3 has suggested that “structure phases” are formed in glass approximating in the case of silicate glasses to mica amphibole and pyroxene Silo ratios and having no relation to the composition of the phases formed on crystallization under equilibrium conditions. It is of interest also to note here that the glass of 5Ca0.3A120 composition has unusual properties showing a refractive index of 1-66 above that of stable 5Ca0.3A1,O3 crystals (1.61) but below that of the unstable crystals (1.69). The effect is not confined to the system CaO -A1,0,. Brownmiller was the first to draw attention to the formation of metastable phases from glasses of compositions likely to be found in quickly cooled Portland cement clinker.Here again it appears possible that a corresponding mineral is sometimes found in thin sections of Portland cement clinker as a prismatic interstitial compound which according to Bogue l5 is unstable and formed only under unusual cooling conditions. Bowen Schairer and Posnjak l6 report an interesting observation on wollastonite. The inversion from high-temperature pseudo-wollastonite to low-temperature wollastonite is so sluggish that once formed it is almost impossible to convert it to the low-temperature form. The latter is therefore normally prepared by annealing glass at below 1150’ C (the inversion temperature).These authors found that on annealing the powdered glass pseudo-wollastonite was obtained whereas annealing a lump of glass l1 Sundius Symposium on Chemistry of Cemmt (Stockholm 1938) p. 393. l2 Dyckerhof Zernent 1924 13 (34) 400. l3 Lukesh Science 1946 104 199 ; Awzer. ,Wirier. 1948 33 76. l4 Brownmiller Amcr. J . Scz. 1938 35 241. l5 Bogue The Chemistry of Portland Cement (Rheinhold Publishing Co. 1g47) p. 132. l 6 Bowen Schairer and Posnjak Amer. J. Sci. 1933 26 207. FIG. 7.-Unstable gCa0.3A1,O3 in high-alumina cement clinker. ( x 50) FIG. 8.-Annealed glass of composition jCa0.3Al2O (crossed polars). ( x 2 jo) [To face page 348 FIG. 9.-Section of twinned p zCaO.Si0 crystal (crossed polars). ( x 50) F. M. LEA AND R. W. NURSE 349 yielded the expected low-temperature wollastonite.These experiments when repeated by the present authors yielded the low-temperature form in both cases. The effect observed by Bowen is probably connected with the incomplete removal of pseudo-wollastonite nuclei from the glass but what- ever the explanation for the beginning of crystallization there seems to be no doubt that once begun the crystallization of pseudo-wollastonite continues at temperatures well below the inversion temperature. The practical problem in building material manufacture is frequently the inverse of that considered here ; namely a study of the glassy condition and its influence in the properties of the product. The effect of glass formation in Portland cement clinker has been discussed by Parker.17 In high alumina cement a low strength is often associated with excessive glass formation.On the other hand in granulating blast-furnace slag for cement production the maximum conversion to the glassy form is desired and this places an upper limit l 8 on the lime content of slags used for this purpose since otherwise crystallization cannot be inhibited by rapid cooling. Recrystallization in the Solid State.-Growt h of new crystals takes place during the decomposition or recombination of compounds to fonn new species on heating by the decomposition on cooling of compounds which are stable only at high temperatures and by inversion of polymorphic foms of one mineral. An example of the former is the formation of 3CaO.Si0 on heating a mixture of CaCO and SiO,.After the decomposition of the carbonate the orthosilicate zCaO.Si0 is first formed even if the temperature is above 1275" C below which aCaO.Si0 and CaO are the stable phases. Tricalcium silicate is formed only after prolonged heating; an atmosphere of steam accelerates the combination. Although crystals of 3CaO.Si0 formed in this way are only a few microns in size good crystal faces are developed. reaction. Below 1275" C it decomposes in the solid state into CaO + The same compound may be used as an example of the second type of aCaO.Si0,. The onset of the reaction is seen as a development of turbidity within the tricalcium silicate grain which is resolved on further heating into birefringent specks of zCaO.Si0 and minute rounded CaO crystals.The reaction takes several hundred hours to complete at 1250" C and even after this time the reaction products are imperfectly crystallized. As might be expected from the high temperature at which they are formed cement and slag minerals frequently show polymorphism. A new form of 3CaO.Si0 has been found by Bernal as a result of X-ray examination of single crystals prepared by the technique developed by Nurse.' The crystallographic relations between the various forms of aCaO.Si0 on inversion have been discussed by Tilley l9 who has also discovered a new high-temperature form. Twinning in such compounds has been related to the polymorphism 2o and Parker and Ryder 21 conclude as a result of an empirical correlation of microscopic structure with dusting of blast-furnace slags that twinning is characteristic of those forms of zCaO.Si0 which are likely eventually to invert to the y forin.Twinning may also presumably arise as a result of stresses imposed by the cooling conditions as well as from inversion. The twinned crystal of zCaO.Si0 shown in Fig. g was grown from melt entirely in the temperature range at which the p form 17 Parker J . SOC. C h e w I n d . 1939 38 203. 18 Parker and Nurse Granulated b2astfwnace slag for cement manufacture. Building Research Technical Paper (H.R.Z. Stationery Office) (in press). l9 Tilley Miner. Mag. 1948 28 255. 2o Inslcy Flint Newman and Swenson J . RPS. Nut. B w . Staitd. 1938 21 355. 21 Parker and Ryder J . Iroiz Steel I~zst. 1942 2 PIP. CRYSTAL GROWTH IN BUILDING MATERIALS 350 is stable but there is no evidence to show whether the twinning existed before cooling the crystal to room temperature.During the inversion of p to y aCaO.Si0 an intemediate condition has been observed when the optical properties are partly those of one and partly of the other phase.,l Such " metaphases " have been reported for a number of solid phase reactions. Eitel 22 has shown by means of electron microscopy and electron diffraction that at any rate in a number of cases there is no true intermediate phase but that nuclei of the new phases form at definite points in the lattice of the decomposing phase giving rise to anomalous optical properties. crystallization of solids in aqueous solutions is an important factor in the cementing of materials and under certain conditions in causing disruption.The setting of plaster can be regarded as one of the simplest cases of growth of crystals which interlock and cement into a solid mass. Though the reaction Crystallization in Aqueous Systems .-The CaSO,.&H,O + r&H,O = CaSO,.zH,O is accompanied by a decrease in volume of over 7 % a mass of plaster expands on setting leaving voids in its interior. The magnitude of the expansion which is influenced by various factors and particularly by the presence of small amounts of other agents usually falls within 0.1 to 1.0 % (linear). There is an initial stage while the mass is still very plastic in which a small contraction appears but this is soon superseded by the expansion which runs roughly parallel to the rate of hydration.It is generally held that in the initial stage before the plaster acquires rigidity the crystals are free to move without restraint but that as soon as a rigid structure is formed unidirectional growth under conditions of restraint causes the observed expansion. Recent unpublished work by Andrews at the Building Research Station has shown that the degree of expansion is related to the crystal habit assumed by the gypsum. With a medium such as water in which the gypsum crystals grow in acicular form a high expansion is normally found (Fig. IO) while in the presence of additions which lead to crystallization of the gypsum in less-elongated and broader forms the expansion is low (Fig.11). The growth of crystals against a unidirectional stress seems to demand more study since in the few recorded measurements the forces developed are generally small. An exception appears in the forces recorded by Correns and Steinborn 23 with crystals of potassium alum but Schubnikow24 for the same case found only very small forces. It is clear 25 on thermodynamical grounds that when a longitudinal compressive stress is applied to a crystal without pressure on the surrounding solution the solubility will be increased more at the stressed face than at the free faces. In the case of the setting of plaster there exists a considerable degree of supersaturation of the solution but no apparent relation between that degree and the form of the crystal growth.Le Chatelier long ago advanced the general theory that cementing action occurred by crystal growth when a system of anhydrous constituents unstable in water reacted to produce a solution which was supersaturated with respect to the stable system of hydrated products. This applies to calcium sulphate plasters and to cements though in the latter case the relative parts played by crystal growth and surface forces of gelatinous constituents has long been subject to controversy. An interesting example of cementing 22 Eitel Prezzssische Akad. Wiss. 1943. Math. Naturw. Klasse No. 5 Berlin 1944. 83 Correns and Steinborn 2. Krist. 1939 IOI 117. 24 Schubnikow 2. Krist. 1934 88 466. 26 Goranson J . Chem. Physics 1940 8 323. FIG. Io.-Elongated gypsum crystals grown in water.( x 250) FIG. I 1.-Stubby modification of gypsum habit by sodium citrate. ( x 250) [To face page 350 F. M. LEA AND R. W. NURSE 351 action resulting from the transformation of a monotropic anhydrous com- pound to its stable form has been cited by Ilchenko and Lafuma.26 The solubility of aragonite in water is slightly greater than that of calcite and a finely ground powder of the former was found to set very slowly with water owing to the growth of calcite crystals. The conversion of one crystalline compound to another can also have destructive effects. The hydration of high alumina cement at normal temperatures leads to the formation of zCa0.A1,0,.8H20 and gelatinous alumina. The hydrated calcium aluminate which is formed as pseudo- hexagonal plate and needle crystals is metastable with respect to the cubic compound 3CaO.Al2O3.6H2O but at 18" it remains stable indefinitely.At higher temperatures (35" - so") the inversion occurs within weeks with liberation of hydrated alumina. Though accompanied by an increase in density and reduction in solid volume the change causes a loss of some 70-80 yo of the strength of the set mass.27 The initial strength development at 45" is also lower than at 18" indicating the poorer binding action of cubic crystals compared with more elongated forms. A similar effect has been found by Andrews with gypsum plasters where the higher strengths are associated with the more elongated forms of the gypsum crystals. As another example of crystal transformation causing disruption there may be cited the action of calcium sulphate solutions (or other solubIe sulphates) on the hydrated alumina compounds present in set Portland cement.The compound 4CaO.AI2O,. 13H,0 or certain solid solutions which it forms reacts with calcium sulphate to form 3Ca0.A1,0,.3CaS04.~~H20. The former compound occurs as hexagonal plates and the latter as elongated hexagonal needles. The increase in solid volume which results is accom- modated not by growth into existing pore spaces but by an outward thrust causing disruption of the solid mass. Lafuma 28 has suggested that the expansion is due to the low solubility of the compounds involved and direct growth from the original hydrated calcium aluminate crystals in situ rather than by passage into and growth from solution.An analogy may be drawn with the high pressures that can be created within a porous material con- taining a saturated solution of sodium sulphate and the anhydrous salt when the temperature is reduced below the transition point of the hydrated decahydrate. 29 These various examples add emphasis to the comment made earlier on the need for more study of the growth of crystals under stress and reinforce the conclusion drawn by Wells that more systematic work is needed on the factors which determine the relative rate of growth of different crystal faces. Buitding Research Station Department of Scient@c and Industrial Research Garston Watford Herts. 26 Ilchenko and Lafuma Chirn. et Ind. 1937 38 438. 27Lea J .SOC. Chern. I n d . 1940 59 18. 28 Lafuma Rev. Mat. Const. 1929 243 I929 ; 1930 244 4. 2 9 Bonnell and Nottage J. SOC. Chcm. Ind. 1939 58 10. L. A. THOMAS N. WOOSTER AND W. A. WOOSTER 345 PROBLEMS OF CRYSTAL GROWTH IN BUILDING MATERIALS BY F. M. LEA AND R. W. NURSE Received 1st February 1949 The growth of crystals is of interest in various directions in the study of building materials and structures. The present paper will be limited to problems arising in the manufacture of building materials and to some crystalline changes which arise subsequently when the material is in use. Crystallization at High Temperatures.-Industrial materials such as slags and cements are produced by sintering or fusing the raw materials at high temperatures and there is considerable technological interest in the extent and manner of crystal growth.With slags and high alumina cement the mix becomes completely molten and crystallization takes place from the liquid phase but with Portland cement only some 20-30 yo of the mix becomes liquid and crystal growth occurs both as a result of solid-liquid reactions and of direct crystallization from the liquid. The study of the phase equilibria diagrams is an essential feature of work on crystal growth in such materials for it enables the order of appearance on cooling of different crystals to be defined and the effect of departures from equilibrium to be traced. The complicated crystallization paths which can arise in polycomponent silicate systems have been discussed by various authors.1 2 They frequently involve disappearing phases and produce typical crystalline structures such as eutectic and peritectic patterns corrodcd crystals and zoning etc.Attention has also been drawn to the possibility of " independent crystallization " of the liquid phase taking place when the cooling process is rapid or involves two or more distinct stage^.^ The final structure in such cases of frozen equilibrium can be predicted from the phase equilibrium diagrams. Crystallization from the Melt.-A most noticeable effect in pofy-component silicate and aluminate systems is the tendency of particular compounds when crystallizing from the melt to appear as spheres (Fig. I). This effect is not related to the symmetry being very obvious in the case of zCaO.SiO, which has quite low symmetry.Optically the crystals Bowen The Evolzttion of the Igneous Rocks (Princeton 1928). 2 Hall and Insley Phase Diagrams for Ceramicists (Amer. Ceram. SOC. 1947). 8 Lea and Parker Phil. Trans. 1934 234 I. Lea and Parker Bzrilding Research Tech. Paper No. 16 (H.M. Stationery Office 1935). Parker and Nurse .J. SOC. Chem. I n d . 1939 58 255. M FIG. 1.-Periclase (MgO) crystals in glass. ( x 250) To face page 345 346 CRYSTAL GROWTH I N BUILDING MATERIALS appear to be single individuals and it is only at a later stage when the crystals are larger or under conditions of slower growth that normal faces and forms appear. I t is generally considered 6 that curved or vicinal faces arise when steep concentration gradients exist in the solution or when the rate of diffusion is low.This can result from a high viscosity of the melt or from a high rate of crystallization. Spherical growth does not occur in melts of high silica content such that glasses are readily formed but it is charac-teristic of compounds such as CaO MgO zCaO.Si0 and 3Ca0.Mg0.zSi02, and to a lesser extent 3Ca0.A12.0 and 5Ca0.3A1203 which either do not form glasses or are easily devitrified. Spinel which readily foms a glass, always develops as minute octahedra (Fig. 2 ) . If instability of the glassy state is to be associated with a high crystallization rate it would appear that the latter as well as the viscosity of the melt is an important factor determining the growth of spherical crystals. Growth of crystals by a solid-liquid reaction does not lead to spherical forms.Thus 3CaO.Si0 formed in this way shows its true symmetry but small crystals grown from a suitable melt are often spherical. Minerals differ much in their rate of crystallization and this frequently leads to complication of the crystallization path and to changes in the form of crystal growth. Where an incongruently melting compound occurs in a system the phase first crystallizing may react with the liquid at a definite temperature to form a new species. Frequently the new crystal grows at such a rate that the dissolving primary crystals are enveloped and excluded from contact with liquid so that on complete solidification the second phase may still contain inclusions of the first. Typical cases are inclusions of CaO in 3Ca0.A1,03 (Fig.3) and MgO in monticellite. If such a formation is held for a sufficient length of time at a temperature just below the saturation temperature of the second phase the inclusions disappear. The rate of diffusion in solid phases at such temperatures must, therefore be quite high. Growth of Single Crystals.-Work on the X-ray structure of cement minerals has been hindered in the past because the low symmetry neces-sitates single crystal determinations and methods for growing the crystals had not been devised. Many of the minerals melt incongruently or as in the case of 3CaO.Si0 decompose below the melting point and cannot, therefore be grown from melts of their own composition by conventional methods. Attempts to encourage crystallization by the addition of mineral-izers have sometimes been successful but many failed because not enough was known of the crystallization paths in complex systems.Le Chatelier, for instance was unable to crystallize 3CaO.Si0 from melt containing CaC1,. and single crystals of pure 3CaO.Si0 up to I mm. and 3CaO.Si0 solid solution up to 5 mm. in length have been grown. Twinned crystals of p zCaO.Si0 up to I cm. long have also been obtained (Fig. 4 and 5). The method used for 3CaO.Si0 is briefly as follows. Fig. 6 is a schematic diagram of the system CaO - zCaO.Si0 - CaCl,. A composition such as denoted by B (35 % 3Ca0.Si02 35.% y zCaO.SiO,, 30 yo CaCl,) was made up from the previously reacted sllicates. A total weight of 5 g. of the mixture was heated in the electric muffle at a temperature of 15ooOC in a platinum crucible 2.5 cm.diam. and 2-5 cm. tall. After a period of about 4 hr. most of the chloride had evaporated; the melt was then cooled in air and " dusted " owing to the @? inversion of aCaO.Si0,. This subject has been taken up again by one of the present authors 6 Wells Ann. Re;borts 1946 84. Nurse 21st Cong. Ind. Chew. (Brussels 1948) FIG. 2.-Spinel (MgO.AI,O,) crystals in glass. ( x 250) FIG. 3.-Inclusions of CaO in 3Ca0.Al,0E. ( x 250) [To fat page 34 FIG. 4.-Single crystals of gCaO.Si0,. ( x 20) FIG. 5.-Twinned crystals of p aCaO.Si0,. ( x 20 F. M. LEA AND R. W. NURSE 347 The large crystals of 3CaO.Si0 were then separated by washing with alcohol on a 300-mesh sieve. Neglecting the decomposition and oxidation of CaCl into CaO and Cl,, the melt composition follows the line BA in Fig.6 and eventually passes into the 3CaO.Si0 primary phase field. Crystals of 3CaO.50 begin to form when the 1500’ C isotherm is reached and since the melt is then losing both CaC1 and 3Ca0.Si02 the melt composition follows the isotherm towards the aCaO.Si0 field. When all the chloride has evaporated the melt has composition A and consists of large crystals of 3CaO.Si0 and zCaO.Si0,. Similar methods have been used to prepare single crystals of a new compound 3SrO.Si0 from the oxides and SrCl,. It is not isomorphous with 3Ca0.Si02.* The successful preparation of spinel boules in the Verneuil furnace suggests that it should be possible to grow single crystals of congruently melting compounds such as CaO.A1,0, rzCa0.7Al,03 ~Ca0.A1,03.Fe,03 by [this method and it is hoped to attempt this shortly.CaO ( 1 5 7 0 ~ ) CaCL2 (7800~) FIG. 6.-System CaO - zC.aO.SiO?- CaC1,. Crystallization from the Glass .-Most studies of crystallization from the glassy state lo have been carried out on ccmmercial glass compositions. In such cases it has been assumed and scmetimes confirmed experimentally, that the phase crystallizing is that to be expected from the relevant phase equilibrium diagram. A number of observations for which there is as yet no connected theory indicates that this is not always the case. In the crystallization of high alumina cement melts a large amount of a phase which is known as ‘ I unstable 5Ca0.3Al2O ” * (Fig. 7) often appears. Rankin and Wright found that both 5Ca0.3A120 and 3CaO.5A1,O3 occurred * Nurse unpublished data.Morey Properties of Glass (Rheinhold Pub. Corp. 1938). l o Morey Trans. Faraduy Soc. 1941 37 209. * There seems to be little doubt that the stable phase identified by Rankin and However Wright in the system CaO - A1,0 is reCa0.7Al,03 and not 5Ca0.3A1,03. for the sake of clarity the older designation will be used 348 CRYSTAL GROWTH I N BUILDING MATERIALS in unstable forms under certain conditions of cooling during their studies in the system CaO -A1,0,. Sundius l1 separated a mineral from high alumina cement clinker which corresponded closely in optical properties to Rankin and Wright’s unstable 5Ca0.3A1,03; the separation was not complete and his analysis showed that the mineral might be 3Ca0.2AI,03.Recent experiments at the Building Research Station indicate a composition higher in lime as being more likely but still not conforming in composition to any compound known in the system CaO - A1,0,. Dyckerhof l2 obtained “ unstable 5Ca0.A1,03 ” by annealing glass of that composition at about IOOOO C. These experiments have been repeated a t the Building Research Station but owing to the dendritic nature of the crystals (Fig. 8) it was not possible to determine whether any glass remained. It was found, however that the “unstable” compound was formed only below 1020’. The crystals in high alumina cement clinker are so well developed that it is difficult to believe that they are derived from devitrification of a glass ; they frequently occur in association with the stable form.A possible explanation is that a low-temperature stable form of 5CaO.-3A1,03 exists or that at low temperatures 5Ca0.3A1,O3 decomposes into CaO.AlqO and 2Ca0.A120, but there is little evidence to support this since crystalline 5Ca0.3Al20 cannot be converted to the unstable form by annealing at any temperature. Furthermore the new species should crystallize from the melt in any polycomponent system in which the liquidus tem-perature fell below the decomposition or inversion temperature of the normal form of 5Ca0.3AI2O,. Liquid temperatures in the system 5Ca0.3A120 -Na,O.WO fall below IOZO’ but the primary 5Ca0.3Al20 still occurs in the stable form.8 There is some solid solution however so there is a possibility that the inversion (decomposition) temperature has been lowered.If these aluminates prove to be truly metastable phases this phenomenon may be of some interest in connection with recent studies of glass structures. Lukeshl3 has suggested that “structure phases” are formed in glass approximating in the case of silicate glasses to mica amphibole and pyroxene Silo ratios and having no relation to the composition of the phases formed on crystallization under equilibrium conditions. It is of interest also to note here that the glass of 5Ca0.3A120 composition has unusual properties showing a refractive index of 1-66 above that of stable 5Ca0.3A1,O3 crystals (1.61) but below that of the unstable crystals (1.69). was the first to draw attention to the formation of metastable phases from glasses of compositions likely to be found in quickly cooled Portland cement clinker.Here again it appears possible that a corresponding mineral is sometimes found in thin sections of Portland cement clinker as a prismatic interstitial compound which according to Bogue l5 is unstable and formed only under unusual cooling conditions. Bowen Schairer and Posnjak l6 report an interesting observation on wollastonite. The inversion from high-temperature pseudo-wollastonite to low-temperature wollastonite is so sluggish that once formed it is almost impossible to convert it to the low-temperature form. The latter is, therefore normally prepared by annealing glass at below 1150’ C (the inversion temperature). These authors found that on annealing the powdered glass pseudo-wollastonite was obtained whereas annealing a lump of glass The effect is not confined to the system CaO -A1,0,.Brownmiller l1 Sundius Symposium on Chemistry of Cemmt (Stockholm 1938) p. 393. l2 Dyckerhof Zernent 1924 13 (34) 400. l3 Lukesh Science 1946 104 199 ; Awzer. ,Wirier. 1948 33 76. l4 Brownmiller Amcr. J . Scz. 1938 35 241. l5 Bogue The Chemistry of Portland Cement (Rheinhold Publishing Co. 1g47) p. 132. l 6 Bowen Schairer and Posnjak Amer. J. Sci. 1933 26 207 FIG. 7.-Unstable gCa0.3A1,O3 in high-alumina cement clinker. ( x 50) FIG. 8.-Annealed glass of composition jCa0.3Al2O (crossed polars). ( x 2 jo) [To face page 34 FIG. 9.-Section of twinned p zCaO.Si0 crystal (crossed polars). ( x 50 F. M. LEA AND R. W. NURSE 349 yielded the expected low-temperature wollastonite.These experiments when repeated by the present authors yielded the low-temperature form in both cases. The effect observed by Bowen is probably connected with the incomplete removal of pseudo-wollastonite nuclei from the glass but what-ever the explanation for the beginning of crystallization there seems to be no doubt that once begun the crystallization of pseudo-wollastonite continues at temperatures well below the inversion temperature. The practical problem in building material manufacture is frequently the inverse of that considered here ; namely a study of the glassy condition and its influence in the properties of the product. The effect of glass formation in Portland cement clinker has been discussed by Parker.17 In high alumina cement a low strength is often associated with excessive glass formation.On the other hand in granulating blast-furnace slag for cement production the maximum conversion to the glassy form is desired and this places an upper limit l 8 on the lime content of slags used for this purpose, since otherwise crystallization cannot be inhibited by rapid cooling. Recrystallization in the Solid State.-Growt h of new crystals takes place during the decomposition or recombination of compounds to fonn new species on heating by the decomposition on cooling of compounds which are stable only at high temperatures and by inversion of polymorphic foms of one mineral. An example of the former is the formation of 3CaO.Si0 on heating a mixture of CaCO and SiO,. After the decomposition of the carbonate the orthosilicate zCaO.Si0 is first formed even if the temperature is above 1275" C below which aCaO.Si0 and CaO are the stable phases.Tricalcium silicate is formed only after prolonged heating; an atmosphere of steam accelerates the combination. Although crystals of 3CaO.Si0 formed in this way are only a few microns in size good crystal faces are developed. The same compound may be used as an example of the second type of reaction. Below 1275" C it decomposes in the solid state into CaO + aCaO.Si0,. The onset of the reaction is seen as a development of turbidity within the tricalcium silicate grain which is resolved on further heating into birefringent specks of zCaO.Si0 and minute rounded CaO crystals. The reaction takes several hundred hours to complete at 1250" C and even after this time the reaction products are imperfectly crystallized.As might be expected from the high temperature at which they are formed cement and slag minerals frequently show polymorphism. A new form of 3CaO.Si0 has been found by Bernal as a result of X-ray examination of single crystals prepared by the technique developed by Nurse.' The crystallographic relations between the various forms of aCaO.Si0 on inversion have been discussed by Tilley l9 who has also discovered a new high-temperature form. Twinning in such compounds has been related to the polymorphism 2o and Parker and Ryder 21 conclude as a result of an empirical correlation of microscopic structure with dusting of blast-furnace slags that twinning is characteristic of those forms of zCaO.Si0 which are likely eventually to invert to the y forin.Twinning may also presumably arise as a result of stresses imposed by the cooling conditions as well as from inversion. The twinned crystal of zCaO.Si0 shown in Fig. g was grown from melt entirely in the temperature range at which the p form 17 Parker J . SOC. C h e w I n d . 1939 38 203. 18 Parker and Nurse Granulated b2astfwnace slag for cement manufacture. l9 Tilley Miner. Mag. 1948 28 255. 2o Inslcy Flint Newman and Swenson J . RPS. Nut. B w . Staitd. 1938 21 355. 21 Parker and Ryder J . Iroiz Steel I~zst. 1942 2 PIP. Building Research Technical Paper (H.R.Z. Stationery Office) (in press) 350 CRYSTAL GROWTH IN BUILDING MATERIALS is stable but there is no evidence to show whether the twinning existed before cooling the crystal to room temperature.During the inversion of p to y aCaO.Si0 an intemediate condition has been observed when the optical properties are partly those of one and partly of the other phase.,l Such " metaphases " have been reported for a number of solid phase reactions. Eitel 22 has shown by means of electron microscopy and electron diffraction that at any rate in a number of cases there is no true intermediate phase but that nuclei of the new phases form at definite points in the lattice of the decomposing phase giving rise to anomalous optical properties. Crystallization in Aqueous Systems .-The crystallization of solids in aqueous solutions is an important factor in the cementing of materials and under certain conditions in causing disruption.The setting of plaster can be regarded as one of the simplest cases of growth of crystals which interlock and cement into a solid mass. Though the reaction CaSO,.&H,O + r&H,O = CaSO,.zH,O is accompanied by a decrease in volume of over 7 % a mass of plaster expands on setting leaving voids in its interior. The magnitude of the expansion which is influenced by various factors and particularly by the presence of small amounts of other agents usually falls within 0.1 to 1.0 % (linear). There is an initial stage while the mass is still very plastic in which a small contraction appears but this is soon superseded by the expansion which runs roughly parallel to the rate of hydration. It is generally held that in the initial stage before the plaster acquires rigidity the crystals are free to move without restraint but that as soon as a rigid structure is formed unidirectional growth under conditions of restraint causes the observed expansion.Recent unpublished work by Andrews at the Building Research Station has shown that the degree of expansion is related to the crystal habit assumed by the gypsum. With a medium such as water in which the gypsum crystals grow in acicular form a high expansion is normally found (Fig. IO) while in the presence of additions which lead to crystallization of the gypsum in less-elongated and broader forms the expansion is low (Fig. 11). The growth of crystals against a unidirectional stress seems to demand more study since in the few recorded measurements the forces developed are generally small.An exception appears in the forces recorded by Correns and Steinborn 23 with crystals of potassium alum, but Schubnikow24 for the same case found only very small forces. It is clear 25 on thermodynamical grounds that when a longitudinal compressive stress is applied to a crystal without pressure on the surrounding solution, the solubility will be increased more at the stressed face than at the free faces. In the case of the setting of plaster there exists a considerable degree of supersaturation of the solution but no apparent relation between that degree and the form of the crystal growth. Le Chatelier long ago advanced the general theory that cementing action occurred by crystal growth when a system of anhydrous constituents unstable in water reacted to produce a solution which was supersaturated with respect to the stable system of hydrated products.This applies to calcium sulphate plasters and to cements though in the latter case the relative parts played by crystal growth and surface forces of gelatinous constituents has long been subject to controversy. An interesting example of cementing 22 Eitel Prezzssische Akad. Wiss. 1943. 83 Correns and Steinborn 2. Krist. 1939 IOI 117. 24 Schubnikow 2. Krist. 1934 88 466. 26 Goranson J . Chem. Physics 1940 8 323. Math. Naturw. Klasse No. 5 Berlin 1944 FIG. Io.-Elongated gypsum crystals grown in water. ( x 250) FIG. I 1.-Stubby modification of gypsum habit by sodium citrate. ( x 250) [To face page 35 F.M. LEA AND R. W. NURSE 351 action resulting from the transformation of a monotropic anhydrous com-pound to its stable form has been cited by Ilchenko and Lafuma.26 The solubility of aragonite in water is slightly greater than that of calcite and a finely ground powder of the former was found to set very slowly with water owing to the growth of calcite crystals. The conversion of one crystalline compound to another can also have destructive effects. The hydration of high alumina cement at normal temperatures leads to the formation of zCa0.A1,0,.8H20 and gelatinous alumina. The hydrated calcium aluminate which is formed as pseudo-hexagonal plate and needle crystals is metastable with respect to the cubic compound 3CaO.Al2O3.6H2O but at 18" it remains stable indefinitely.At higher temperatures (35" - so") the inversion occurs within weeks with liberation of hydrated alumina. Though accompanied by an increase in density and reduction in solid volume the change causes a loss of some 70-80 yo of the strength of the set mass.27 The initial strength development at 45" is also lower than at 18" indicating the poorer binding action of cubic crystals compared with more elongated forms. A similar effect has been found by Andrews with gypsum plasters where the higher strengths are associated with the more elongated forms of the gypsum crystals. As another example of crystal transformation causing disruption there may be cited the action of calcium sulphate solutions (or other solubIe sulphates) on the hydrated alumina compounds present in set Portland cement.The compound 4CaO.AI2O,. 13H,0 or certain solid solutions which it forms reacts with calcium sulphate to form 3Ca0.A1,0,.3CaS04.~~H20. The former compound occurs as hexagonal plates and the latter as elongated hexagonal needles. The increase in solid volume which results is accom-modated not by growth into existing pore spaces but by an outward thrust causing disruption of the solid mass. Lafuma 28 has suggested that the expansion is due to the low solubility of the compounds involved and direct growth from the original hydrated calcium aluminate crystals in situ rather than by passage into and growth from solution. An analogy may be drawn with the high pressures that can be created within a porous material con-taining a saturated solution of sodium sulphate and the anhydrous salt when the temperature is reduced below the transition point of the hydrated decahydrate. 29 These various examples add emphasis to the comment made earlier on the need for more study of the growth of crystals under stress and reinforce the conclusion drawn by Wells that more systematic work is needed on the factors which determine the relative rate of growth of different crystal faces. Buitding Research Station, Department of Scient@c and Industrial Research, Garston Watford Herts. 26 Ilchenko and Lafuma Chirn. et Ind. 1937 38 438. 27Lea J . SOC. Chern. I n d . 1940 59 18. 28 Lafuma Rev. Mat. Const. 1929 243 I929 ; 1930 244 4. 2 9 Bonnell and Nottage J. SOC. Chcm. Ind. 1939 58 10

ISSN:0366-9033    

DOI:10.1039/DF9490500345

出版商:RSC    

年代:1949

数据来源: RSC

摘要:

THE GROWTH OF PERICLASE CRYSTALS AND ITS IMPORTANCE IN BASIC REFRACTORIES BY E. B. COLEGRAVE, H. M. RICHARDSON AND G. R. RIGBY Received 7th February, 1949 Although crystal growth is regarded as an important factor influencing the mechanical properties of ceramic bodies, there appear to be few, if any, investigations which have been made specifically to study this point. It has been suggested that a high mechanical strength in certain porcelain bodies is due to the growth of mullite crystals which can interlock owing to their acicular habit. In general, however, crystallization from a glass phase is accompanied by a deterioration in mechanical properties, and the larger the individual crystals, the more marked is the reduction in strength. This latter point is generally appreciated in the fabrication of high-temperature oxide materials by the process of sinterhg, the problem being to reduce the pore spaces while avoiding marked crystal growth.There is one refractory material, however, magnesia, which is preheated before use in order to encourage large crystals to develop. Commercial " magnesite " when made into bricks or blocks finds important industrial applications in the linings of basic steel furnaces, copper-refining furnaces, rotary cement kilns and metal mixers for pig iron. It is obtained by calcining either the natural carbonate rock or magnesium hydroxide obtained by decomposing the magnesium salts present in sea water. During this calcination the carbonate or hydroxide is decomposed to the oxide at temperatures below 800" C, but this so-called caustic magnesia is unsuitable for refractory uses, as bricks made from the material would possess excessive volume- shrinkage on being subjected to steel-making temperatures ; further the bricks would hydrate fairly rapidly on exposure to the atmosphere which would cause them to crumble. The magnesia of the refractories industry has therefore to be calcined to a temperature of around, or above, 1600OC before being pressed into bricks, this high-temperature calcination being referred to as dead-burning.It has been known for many years that the dead-burning obviated excessive shrinkage and hydration tendency, the process generally being accompanied by a marked increase in specific gravity as determined by the conventional density bottle method.This increase in specific gravity between calcining at 1300OC and 1600OC was at one time thought to be due to the oxide altering its crystalline form, but X-ray analysis has subsequently shown that this is not the case, the effect of increasing the calcination temperature being to increase the size of the individual periclase crystals, the lattice remaining unchanged. Neverthe- less, the specific gravity determination is often used as a criterion of the degree of dead-burning which the material has received. It has been known for many years that impure magnesite rock is more readily dead-burnt than the purer carbonates. The magnesite rocks from Styria are of the breunnerite type and contain from 5 yo to 30 yo of ferrous carbonate in solid solution with the magnesite.These sources supply an excellent dead-burnt oxide, whereas the purer Grecian magnesites are more difficult to dead-burn and the resulting oxide may still possess a residual shrinkage. The effect of impurities on the rate of crystal growth of the periclase crystals is a problem having practical applications, and when Germany, before the 3 5 2FIG. I .-Rounded periclase crystals in a magnesia grain. ( x 40) FIG. 2.-Periclase crystals showing the disappearance of boundaries and the development of cleavage. ( x 40) FIG. 3.-The small crystals have lost their FIG. 4.--Periclase crystals from the identity and cleavage has developed cooler face of a brick after use. in two directions. ( x 40) ( x 66:) FIG. 5.--Periclase crystals from the FIG. 7.--Periclase crystals with numerous working face of a brick after use.small inclusions of magnesioferrite. ( x 663 ( x 106) To face $age 3531E. B. COLEGRAVE, H. M. RICHARDSON AND G. R. RIGBY 353 annexation of Austria in 1936, banned the import of Austrian magnesite, patents were taken out to encourage crystal growth of the purer magnesites by additions of calcium and magnesium ferrite. The sizes of individual periclase crystals in magnesite refractories can be readily measured by examination of a thin-section under the microscope. Chesters, Clarke and Lyon found that the average diameter of periclase crystals measured by this means agreed well with the size obtained from a study of X-ray back-reflection photographs. These authors found that various commercial magnesite bricks contained crystals of an average diameter 0.6 to 0.8 mm.Later Jay and Chesters,, examining a range of magnesite bricks, found a wide variation in crystal size, although in any given brick the size was fairly uniform. In such bricks the periclase crystals are usually rounded as shown in Fig. I but after receiving more severe heat treatment one or more cleavages are developed, the crystals at the same time increasing in size (Fig. 2). Finally, the crystal may show well-defined cleavage cracks in two directions as shown in Fig. 3. The significance of the crystal growth and development of cleavages which usually occurs at the face of a magnesite brick exposed to severe temperatures over long periods as well as to the action of slags and vapours is not known with certainty.have suggested that this crystal growth may lower the resistance of the material to withstand sudden temperature changes without cracking although the development of cleavages may have the reverse effect. Fig. 4 and 5 illustrate the growth which can occur in the individual periclase crystals composing a magnesite brick during service; the periclase crystals in Fig. 4 taken from the colder face of the brick after service are probably of the same order of magnitude as those in the brick before use, but it can readily be seen from Fig. 5 that the periclase crystals at the hot face have increased appreciably in size and have developed a distinct cleavage. The Effect of Temperature and Impurities on Crystal Growth.- Measurements of the variation in crystal growth of periclase crystals with calcining temperature have been made by several investigators.Letort and Halm 3 selected a product from sea water containing 2.5 yo CaO, 0.5 yo SiO,, 0.5 yo Fe,O, and 0.10 yo Al,O, and measured the crystal growth by use of the microscope. A general increase in crystal size was noted as the calcination temperature was raised from 1600" to 1800~ C. Growth also occurred on increasing the time of heat treatment, as on maintaining a temperature of 1800" C for one hour the crystals increased in length from 10-15 p to 35-40 p. The effect of impurities in promoting crystal growth has also received attention, Letort and Halm found that additions of 5 yo Fe,O, and 3 yo SiO, to the sea-water magnesia resulted in periclase crystals 60-70 p and 50-60 p in length respectively on heating for one hour at 1600~ C, whereas if no additions were added the crystal size remained at 25-30 p.These investigators also observed that a reducing atmosphere promoted crystal growth. The importance of kiln atmosphere on the structure of magnesite refractories was also emphasized by Krause and Ksinsik 4 although they consider that, in the dead-burning process, atmo- sphere is only of secondary importance. has used X-ray methods to estimate crystal size, and the graph Fig. 6 shows the variation in crystal size with temperature for three different magnesites. Sample I had a very Jay and Chesters Jay 1 Chesters, Clarke, and Lyon, Trans. Brit. Cevam. Soc., 1935, 34, 243. 2 Jay and Chesters, Trans. Brit. Ceram.Soc., 1938, 37, 218. 3 Letort and Halm, Chim. et Ind., 1947, 58, 537 Krause and Ksinsik, Feuerfest, 1932, 8, 6. Jay, J . Sci. Instr., 1941, 18, 81.354 THE GROWTH OF PERlCLASE CRYSTALS low content of impurities while sample 3, on analysis, contained 5 yo of oxides other than magnesia. Tentative theories have been suggested from time to time to try and explain the mechanism by which certain impurities foster crystal growth. Both Letort and Konopicky consider that ferric oxide is combined as magnesioferrite which is soluble in the periclase crystals at high temperatures, although this may be precipitated on cooling. When magnesite refractories are exposed for considerable periods to ferruginous slags, the surface of the brick may consist of magnesioferrite with magnetite in solution, but immediately behind this layer it is common to find large periclase crystals containing minute inclusions of magnesioferrite, which have presumably been deposited from solution on cooling (Fig.7). Periclase and magnesio- ferrite both crystallize in the cubic system though the cell size of MgO is only half that of the spinel. Tanaka 7 found that titanium dioxide additions ,010, Te mpetatu re OC FIG. 6. also facilitated the sintering of magnesia, and he attributed this to the formation of magnesium orthotitanate which also has a spinel structure. Theories based on solution effects promoting crystal growth, however, cannot be used to explain the mineralizing effect of silica and lime, as it is unlikely that these oxides form minerals which are soluble to any extent in the periclase crystals, and Krause and Ksinsik have stated that the effect of iron oxide on the dead-burning process is less than that of lime or silica.It is, of course, possible that at high temperatures, iron oxide is in solution in the periclase crystals as FeO since FeO and MgO are com- pletely miscible in all proportions in th2 solid state, and that the marked defect structure of FeO might promote crystal growth. I t has been shown Konopicky, Bey. dt. Keram. Gcs., 1937, 18, 97. Tanaka, J . SOC. Chem. Ind., Japan, B, 1939, 42, 202.E. B. COLEGRAVE, H. M. RICHARDSON AND G. R. RIGBY 355 that at high temperatures magnesioferrite is partly decomposed owing t o loss of oxygen. The effect of a reducing atmosphere in enhancing crystal growth could be explained by any theory which postulated the partial dissociation of magnesioferrite to FeO followed by its reformation on cooling, and it is probable that measurements of oxygen pressures in the MgO-FeO- Fe203 system would provide useful confirmatory data.Recent data obtained by the authors 011 the effect of growth of the periclase crystals with (a) temperature and (b) additions of ferric oxide are given graphically in Fig. 8. The magmesite was a natural rock which on calcination contained only 3.56 yo SO,, 0.40 % Fe,O,, 0.74 % Al,O,, 0.66 o/o CaO and 2-12 % alkali metal oxides as impurities. The ferric oxide additions were introduced as 0.005 0.004 E W N u7 - 0 002 - 0 +J Y) 2 v 12 ~ I I 1 3 1300 I400 1500 16C Tempsrature OC FIG, 8. ferrous oxalate which was dry-ground with the magnesite rock after a preliminary calcination a t 800' C.The results showed that a large increase in crystal size was observed as the calcining temperature was raised from The Effect of the Growth of Periclase Crystals on (1) the Volume Shrinkage, (2) the True Specific Gravity and (3) the Hydration Tendency .-The authors have investigated the volume shrinkage undergone by magnesia on heating at various temperatures up to 1600' C by measuring the diameters of standard cylinders made by compressing the powdered magnesia calcined to 800' C. The effect of additions of Fe20, up to a maximum amount of 5 yo was also studied and the results given in graphical 1350' c to 1450' c.356 THE GROWTH OF PERICLASE CRYSTALS form in Fig.9. It may be observed that maximum shrinkage occurred over the range 1350" to 145o"C, but the specimens were still shrinking at 1600~ C. The true specific gravities as determined by displacement methods of various commercial dead-burnt magnesites may vary from 3.56 to 3-65. Jay and Chesters have emphasized that the crystal size, and therefore the degree of dead-burning of the magnesite, bore no simple relation to the specific gravity unless a correction was first made to allow for the iron oxide content of the brick. When using specific gravity values as a criterion of dead-burning it is usual to assume an increase of 0.01 in the specific gravity for each per cent. of Fe,O, in the magnesite analysis. According to calculations made from X-ray photographs the theoretical density of I 1 I 1 2 0 0 1309 I 4 0 0 1500 I608 FIG.9. Temperature c pure magnesia is 3.58 and this does not alter with the temperature of calci- nation. In the above samples, the variation in specific gravity with the temperature of calcination as determined by density bottle measurements ranged from 3.490, with no addition on calcining to 1200" C, to 3.607 with the addition of 5 yo Fe,O, after a calcination to 1600" C. These data can only be reconciled with the X-ray results by assuming that at low tem- peratures of calcination the magnesia contains micro-pores to which the penetrating liquid used in the densitv bottle cannot gain access. Micro-pores enable lightly calcined magnesia powder to be used as a thermal insulator, and their disappearance on increasing the calcining temperature is accom- panied by an overall volume shrinkage.The relationship between calcination temperature and hydration tendencyE. B. COLEGRAVE, H. M. RICHARDSON AND G. R. RIGBY 357 is given by the curves in Fig. 10. The hydration tendency is measured by exposing the finely powdered magnesia to steam at IOO'C for 5 hr., drying finally to 110' C and determining the loss in weight on igniting the sample. It will be observed that a sharp decrease in the hydration tendency occurred as the calcination temperature was raised from 1300' to 1400' C. Conclusions.-It will readily be appreciated that previous investigations on the changes taking place during the dead-burning of magnesia have been made mainly with the practical object of obtaining a better product at lower cost. Among the more theoretical aspects which require further 2 5 2c x 15 U C u V C W I- 10 C 0 w 0 L 0 2. .- r 0 s 0- I 2 0 0 1300 1400 I 5 0 0 1600 Temperature JC FIG. 10. elucidation are the mechanism by which crystal growth occurs and is promoted by certain mineralizers, and the significance to be attached to the development of cleavage planes on subjecting periclase crystals to high temperatures for long periods. The authors are grateful to the Director of the British Ceramic Research Association, Dr. A. T. Green, O.B.E., for permission to publish this paper. British Ceramic Researclz Association, The Mellor Laboratories, Hanley, Stoke-on-Trent.

ISSN:0366-9033    

DOI:10.1039/DF9490500352

出版商:RSC    

年代:1949

数据来源: RSC

摘要:

GENERAL DISCUSSION Mr. Y. Haven (Eindhoven) said : Several authors reported that impurities may be useful in making good crystals. We, too, have made several good, single LiF crystals from a melt to which MgF, (up to a few 0.1 %) was added, in contradistinction to Prof. Stockbarger’s experiences who stated that the I i F should be of the utmost purity, The influence of the impurities may be discussed in the following way. On the one hand, for lattice defects in ionic crystals (in most cases, vacant lattice sites) the law of mass action should be approximately valid, i.e., the product of the concentrations of vacant Lif sites and vacant F’ sites in LiF should be a constant so that one may decrease the number of vacant F’ sites by increasing the number of vacant L+ sites (e.g., by substituting 2Li+ ions for I Mg+++ I vacant site).On the other hand, one kind of ion may often take a dominant part in building up an ionic lattice, thus the negative ions often form the lattice framework, the positive ions having only a more or less supplementary function. This mani- fests itself by the fact that i t is often far more difficult to replace the negative ions than to replace the positive ions. Now defects in the lattice of the framework ions may be more harmful for growing good crystals than the defects of other lattice sites. Thus in LiF one has to prevent defects in the F- lattice, which may be accomplished by adding Mg++ ions. Mg++ ions (i.e., vacant Li+ sites) may therefore have a healing effect on the F- lattice in LiF. For the same reason, i t is possible that one must not push the purification too far, for by removing the positive ion impurities one may introduce negative ion defects.The LiF is mixed with an excess of NH,F. On heating, the H,O is evaporated first and then the excess of NH,F. In this way hydrolysis does not give any trouble, even when the crystal is in contact with air. I would also like to ask why Prof. Stockbarger prefers the crucible-lowering method to that of Kyropoulos, since the latter has the advantage of easy manipulation and inspection and the temperature gradient is such as to cause a good circulation in the melt, which may be important for the growth of the crystal. Dr. W. Ehrenberg and Mr. J. A. Franks (London) said : We obtained good results in the preparation of single alkali halide crystals by using a modification of Stober’s technique.Our furnace has a high thermal inertia ; the main heaters at the top and bottom consist of Kanthal strip and wire embedded in ceramk. No mains stabilization is used. The temperature and temperature gradient are measured with Pt-PtRh thermocouples permanently inserted in the melt ; this appears to be the only way of obtaining information about the actual conditions. Single crystals are almost invariably obtained, even with flat-bottomed crucibles, provided a positive temperature gradient (top a t the higher tempera- ture) is maintained throughout the crystallization. No crucible material mas found to which the crystals do not adhere ; therefore, in order to avoid strains being introduced, it is necessary to use very flimsy containers : our crucibles are made simply by bending a circular sheet of I/IOOO” Pt foil around a former resembling a cog-wheel ; the crucibles look like fairy-cake tins about 24” diam.and I” high. This peculiar shape does not interfere with the growth of the crystal. The crystals are partially orientated with respect to the flat bottom (100 axis parallel to bottom) ; an orientation with respect to the temperature gradient cannot be expected for alkali halides ; in general, a careful study of the literature gave us no evidence that such an orientation has ever been observed. In all cases some Pt is dissolved in the melt and collects a t the top. The bulk of this appears as a dirty irregular deposit on the top of the crystal, while the bottom half is pure ; the top half appears cloudy under u.-v.light. One KC1 crystal grown in an Au crucible beautifully demonstrates the distribution of impurity (Au), the crystal being quite colourless near the bottom and turning deep purple towards the top, on which specks of a thin metallic gold deposit are visible. We can observe no strains in crystals cleaved by tapping with a blunt tool. Strains appear, however, in crystals cleaved with a chisel. I would recommend the use of XH,F in making LiF single crystals. 358GENERAL DISCUSSION 359 Mr. P. R. Rowland (London) said : From what Dr. Ehrenberg has said, there appears to be no orientating effect due to the thermal gradient in the growth of alkali halides from the melt. This is, however, not always the case with metals.Prof. G. I. Finch informs me that in the case of zinc, which has a close-packed hexagonal structure, it is difficult to grow a single crystal unless there is only a small angle between the line of steepest descent of the temperature and the cleavage plane of the crystal. However, in the case of copper, which is cubic close-packed, there appears to be no such effect. We have grown many and have not been able to observe i t (though the number is insufficient to justify a statistical analysis). Neither have we found any reference in the literature to such an orientating effect. There may be an exception to this in one of our experiments, where a thin strip was grown from the melt. The crystal formed had cube direction a t an angle of r8go to the surface plane of the strip. About I$ cm.from the lower (first formed) end of the crystal a second one about IB mm. wide appeared approximately in the centre of the strip, which was I cm. wide. This second crystal had a cube direction a t an angle of only 8” to the surface. The two shared a common cube direction which lay in the surface plane and was normal t o the direction of growth. Is i t possible that this second crystal appeared because there is a tendency to orientate with a cube direction parallel to the direction of growth ? Since the strip was thin (0.5 mm.) an alternative is that surface energy effects may tend to orientate with a cube plane in the surface. Mr. P. R. Rowland (London) (communicated) : Since making the above remarks, the author has discussed the matter with Dr.W. Willman and Prof. G. I. Finch and finds that he was under a slight misapprehension. Zinc crystals grow most quickly parallel to the cleavage plane, and hence a seed crystal orientated with this plane nearly parallel to the length of a rod-shaped melt will tend to outgrow others. There is thus no evidence that the thermal gradient has any effect in orientating the seed in the growth of either zinc or copper from the melt. Mr. T. A. Kletz (I.C.I., Billingham) (communicated) : I n their paper Dr. Menzies and Dr. Skinner state that the discovery of the transmission of silver chloride in the infra-red is probably to be ascribed to Dewar. Actually the discovery was made by Schulze-Sellack nearly 80 years ago,’ later two of the early workers on infra-red spectroscopy, Rubens and Nichols, used silver chloride for the window of their radiometer, taking measurements a at wavelengths up to 24 p.Mr. A. E. Robinson (B.N.S.S.) said : My paper describes plant and apparatus in use for growing small supplies of single crystals for development purposes. The principles do not differ from those in use by a number of workers in this field. A similar ammonium dihydrogen phosphate plant was operated by the Ministry of Supply for the Admiralty before this unit was assembled. In single crystal growing a main concern is the severely practical problem arising from what Dr. Holden has referred to as “maintaining a metastable system for long periods of time.” Of general interest, however, is the fact that the work affords opportunities for observing considerable numbers of crystals of a size and quality not usually met with, and points which may be missed in smaller crystals become obvious.The first is that rate of growth depends on quality. Under given conditions crystals free from obvious flaws do grow a t a reasonably constant and uniform rate ; seeds with inherent flaws which persist grow a t higher rates and the most flawed crystals, but which still retain geometric identity, grow at the highest rate, which is three to four times as great as the slowest-growing sound crystals. These are observations on crystals grown in solutions free from significant impurities. I n general foreign substances in the solution fall into three groups : those without significant effect, e.g., I yo or 2 yo of sodium sulphate in lithium sulphate solutions ; those with specific effect, and here it must be remembered 1 Schulze-Sellack, Pogg.Ann., 1870, 139, 192. 2 Rubens and Kichols, Ann. Physik, 1897, 60, 418 ; and Baly, Spectvoscopy, 3rd ed., vol. I (Longmans, 1g24), p. 230.360 GENERAL DISCUSSION the amounts required may be as small as zo parts per million or less ; and those which appear to encourage the required sound growth, e.g., the small amounts of iron added to ammonium dihydrogen phosphate solution. Dr. E. W. Fell (Bradford) said : Dr. Holden and Mr. Robinson describe a method of growing crystals by moving them through the surrounding solution. However, I should like to refer briefly to a mode of growth in melts when the melt is in motion, When a melt flows along a solid surface such as a mould wall, and under conditions whereby heat is rapidly conducted from the melt through the mould wall, crystallization proceeds from the surface and long crystals grow into the melt, and these crystals, of which the solidified layer near the mould wall is composed, are found to be inclined towards the direction from which the stream is coming.If there is no stream, the crystals grow approximately normally to the mould wall. Such growth is of interest in metallurgy.s It has been observed in steel ingots, lead and stearine. For steel ingots the inclination of the crystals to the normal to the mould is about IO', for stearine about 14' and for aluminium containing 10 yo magnesium (poured down one side of the mould) as much as zoo.The effect on the inclination of different velocities of the stream was not investigated, though there was probably a strong forced circulation of melt in the mould in the case of the aluminium alloy. It is reported that growing crystals of electro-deposited nickel are similarly inclined if the electrolyte is in motion. I have no information regarding crystallization from a solution in motion, but the method of growing crystals described by the authors suggests that a somewhat similar inclination as for melts may occur. It seems that the stream promotes nucleation and the removal of barriers to growth on that boundary of the growing crystal facing the oncoming stream of melt and hence the crystal grows more there, whereas on the other side of the crystal that is sheltered from the oncoming stream there is less nucleation and barrier removal.I should be glad to have the authors' views as to the cause of this observed inclination. Mr. A. E. Robinson (R.N.S.S.) said: The phenomena reported by Dr. Fell occur in crystallization from melts, whereas my paper refers to the somewhat different condition of growth from solutions. The following points may, however, be of interest. In crystal growth in solution there is a tendency for increased growth on the face normal to the stream, and in practice the growing crystal is deliberately oriented so as to encourage growth in the required direction. In spontaneous nucleation, such as that reported on the cold mould wall, there appears to be a tendency for a particular face to adhere, and this is somewhat similar to my exhibit of extraneous growth on selected faces.The angles adopted by different materials are probably influenced by the morphology and habit of the crystal, the wetting of the mould wall and the fact that in a given crystal individual faces have their own degree of wetting by particular fluids. Dr. B . Raistrick (Birmirtgham) said : When sodium trimetaphosphate crystallizes from a melt bubbles of what is believed to be moisture can be seen escaping from the crystal surface as growth proceeds; the loss is weighable. We believe that this observation can be explained on chemical grounds, but would be interested to hear of any other cases of melts absorbing moisture which is then liberated on crystallization.Prof. R. M. Barrer (Aberdeen) said : I n my paper I have given a selection from the many factors which govern the growth of crystalline silicates. The discussion centres round methods of synthesis, crystal dimensions and some of the variables which control growth, including functions of mineralizers. A big literature has grown up in mineral chemistry, but in this country chemical aspects of mineral growth have failed to attract attention in the same way as some better-known phases of chemical research. Much exploratory work mas done long ago-perhaps among the first published silicate syntheses one may name Schafheutl's preparation of quartz in 1845.5 On the Continent there has Carlsson and Hultgren, Jernkontorets Ann., 1936, IZO, 577. Munchner gelehrte Anzeigen, 1845.557. * This Discussion.GENERAL DISCUSSION 361 been continued activity over a long time, and since the turn of the century this is also true of the United States. We owe to Morey and his colleagues in America many of the principal quantitative measurements using the hydrothermal technique. Next to compounds of carbon those of silicon are among the most numerous. Silicates also comprise in bulk a large part of the lithosphere, but techniques for growing silicate crystals are unusual, often falling outside the range of ordinary chemical experience. The crystals are mainly of the " giant molecule " type, containing very large anionic networks, corresponding to chains, sheets and three-dimensional frameworks, and most usually grow from magmas of high viscosity a t high temperatures.Many of the crystals transgress the law of constant proportions due to isomorphous replacements such as K+ + Ka++ : 2Na+ + Ca++ : Xa+Si+++ + Ca++Al+++, and optical, X-ray and chemical data may all be required to establish identity of species. Chemists working in this field naturally depend upon mineralogy and geology in the first place to show the conditions which are likely to yield some a t least of the various species. 1300 I200 I100 loo0 y 900 e 800 $700 Y a 5 600 500 400 300 200 I00 I- 10 20 30 40 50 60 K>Si*O$ WEIGHT PERCENT LEUCITE KAISI,O~- FIG. I. Nevertheless synthetic silicate chemistry will no doubt move in different directions and will reveal many species and types of behaviour not hitherto observed in nature.To some extent this is already true. Thus in my own experience, either by using standard methods or by developing new procedures, crystals have been grown apparently as yet not noted in nature. As examples of new syntheses by standard methods there are the growth by the hydrothermal technique of a new barium zeolite with chabazite-like sorptive properties and of several intermediate species containing the constituents BaCl,, BaBr,, KC1 and KBr in solid solution throughout aluminosilicate frameworks. Dr. Taylor a t my suggestion has by the same method successfully crystallized some thallous aluminosilicates, and Mr. White and also myself have grown a number of lithium aluminosilicate crystals which do not so far appear t o have naturally occurring counterparts. As an example of new methods one may mention the Clark and Steiger procedure for easy production of a number of ammonium minerals by ion inter- change using NH&1 vapour.The NH,+ ion is not found in natural alumino- silicates. I have been able to develop this procedure further in several cases by burning out the ammonium ions with O? gas to form crystalline hydrogen chabazite and mordenite. In natural conditions hydrogen zeolites are never found.362 GENERAL DISCUSSION Again by indirect methods one may easily obtain species which cannot be grown directly under the same conditions of pressure and temperature. Prof. Wyart reports that he has not been able to grow leucite directly by hydrothermal methods ; however, I have made i t easily by first growing analcite hydro- thermally and then submitting this species to the cation interchange, Ka.+ H,O + K+.In this preparation, reaction goes so easily that i t is not necessary to work appreciably above 200' C whether in analcite synthesis or in ion interchange. Clearly synthetic methods are destined to extend and diversify knowledge given by the natural reactions of mineral chemistry. With regard to mineralizers, perhaps water is the most universal, and in con- nection with its mode of action I wish to show an additional diagram containing results very recently published by Tuttle.6 I noted that water may act not only by lowering the viscosity of melts, but also by lowering the crystallizing tempera- tures from the magma. This mechanism is especially important in growth of feldspathic crystals which somehow occurs very well a t rather low temperatures from magmas which if anhydrous would be of astronomical viscosity. Fig.I shows the fusion curves of K,Si,O, (or quartz), orthoclase and leucite in equili- brium with anhydrous and hydrous magmas. Curve I is for an anhydrous magma ; curve 2 for a magma under a water pressure of 15,000 lb. in.-z (i.e., 2.3 miles deep) ; curve 3 for a magma under water pressure of 30,000 lb. in.+ (i.e,, 4.6 miles deep), The fusion temperature of KA1Si,08 is considerably lowered as also is that of leucite. Water a t these pressures actually eliminates growth of K,Si,09 ; instead, quartz appears a t a temperature 300' C lower. Prof. W. E. Garner (Bristol) said : In order to account for the effect of mineralizers in the crystallization of quartz, i t is possible that the mineralizer facilitates the crystallization a t the repeatable step by increasing the mobility of the silica molecules over the quartz surface or in the adjacent liquid phase.Dr. G. R. Rigby (Stoke-on-Trent) said : Dr. Barrer has mentioned that leucite has not yet been successfully synthesized by hydrothermal methods-no doubt Dr. Barrer knows that leucite can be synthesized readily by heating the requisite proportions of potash, alumina and silica and it is often found in used blast- furnace linings where firebricks have been exposed to potash vapour. This arti- ficial leucite exhibits all the characteristic properties of the natural mineral, e.g., the polygonal form, low birefringence and cross-hatched twinning.With regard to Dr. Van Praagh's paper I am surprised that he has identified the high-temperature form of cristobalite a t room temperatures. Cristobalite is the stable modification of silica above 1470" C, but in practice it is often obtained by exposing quartz or fused silica to temperatures above 870" C. If, however, quartz is heated under molten sodium chloride, tridymite is obtained, thus illustrating the specific effect of mineralizers. The inversion of high to low cristobalite is accompanied by an increase in volume amounting to over 3.0 yo, and this is often detrimental to ceramic materials containing the mineral. If the high-temperature form could be stabilized, thus inhibiting this inversion, it would mark a great advance in ceramic technology. Prof.R. M. Barrer (Abevdeen) said : Questions have been asked about the growth of garnet and the functions of mineralizers, and the appearance of leucite in the glass-making furnace has been commented on. Pyrolytic syntheses of leucite are very common, and it is unnecessary to attempt to summarize them. One is in no way surprised a t its appearance during glass-making operations. What is still doubtful, however, is its growth by direct hydrothermal methods a t low temperatures. has not succeeded in repeating Friedel's claims, and although I have carried out direct hydrothermal crystallizations of potassium aluminosilicate gels of varied compositions to give diverse species, leucite has not so far been noted among them, a t least up to 360' C. On the other hand, by the indirect hydrothermal route, already referred to in my paper, leucite has been very easily made a t temperatures of cu.200' C. Thus Prof. Wyart 6 Amer. J. Sci., 1948, 246, 31. 7 This Discussion.GENERAL DISCUSSION 363 A detailed mechanism cannot a t present be given for the growth of garnet in metamorphic conditions. Xevertheless its appearance under high pressures is favoured by its large density according to thermodynamic principles. The symmetrical growth suggests a plastic flow or softening of neighbouring crystalline species of lower density also under the great pressure and in contact with the growing garnet nucleus or crystallite. At the surface of contact the chemical constituents of the other species are then reorganized under stress, so as to decrease the volume occupied and relieve the stress, by continuing the develop- ment of the garnet.In discussing the action of mineralizers a number of speculations have been made. One should not, however, in devising special mechanisms, forget the quite normal aspects. These are that the mineralizer may lower the viscosity of the medium and so promote mixing and crystal growth; that i t may alter the solubility and fusion temperatures of crystallizing species; and that i t may form intermediate compounds. There is good evidence that examples of all these effects occur in various instances. Mr. R. W. Nurse (D.S.I.R., Watford) said : Since our paper was written new information has come to hand concerning some of the examples cited. Sirota 8 has discussed the crystallization of metastable phases, particularly in binary metal alloys, using the Volmer-Stranski method for obtaining the work of forma- tion of two- and three-dimensional nuclei.The examples given show that in general there are three temperature domains : a high-temperature region in which only the stable phase crystallizes, a lorn-temperature region in which only the metastable phase crystallizes, and an intermediate range in which the phase crystallizing depends on the kind of nuclei present. The theory gives a qualitative explanation of the behaviour of the unstable aluminates and also explains the continued growth of pseudo-wollastonite in the wollastonite field as observed by Bowen. Tromme1,e by means of X-ray studies in the high-temperature camera, finds that pzCaO.SiO,, previously thought to be a high-temperature modification, is stable only a t low temperatures. In the first cycle he obtains the inversion y + a’ on heating to 1000’ C and a‘ + p on cooling ; the second cycle gives p -+ z’ (heating) and a’ -+ p (cooling) ; during the third cycle a new modification p’ appears on cooling. This work requires confirmation and extension, but the results already obtained would explain why the crystals of pzCaO.Si0, grown for X-ray structure work have always shown inversion twinning as shown in Fig.g of our paper. The successes obtained with the Verneuil technique reported by Zerfoss l e are most encouraging. When using the “ eutectic ” method of crystallization, where the large number of components used in the melt often prevents any adequate forecast of the phase relations being made, it is particularly necessary to report chemical analyses of the resulting crystals.For instance, in the case quoted by Zerfoss, it seems very likely that a solid solution of BaTiO, and Ba,Ti,O, might be obtained, as is the case with the corresponding calcium compounds. This paper reviews an ingenious process for growing quartz which may offer an improvement in growth rate and quality over that by which rock crystal was produced in nature. A modification of the Spezia method, however, using crystalline quartz as nutrient supply a t an elevated temperature and a thermal gradient such that the region about the seed is a t a lower temperature, has been used with considerable success at The Brush Development Co., Cleveland, Ohio.ll More than half an ounce of quartz, free from cracks and veils, has been deposited on an untwinned R-plate having an area of about 7 cm.2 on a side. Contrary to the recommendation in the paper under discussion, the formation of spontaneous crusts on the wall of the chamber has been avoided as far as possible. The presence of such a coating, presenting a large area of rhombohedra1 quartz on m-hich quartz can deposit from the supersaturated solution, may be helpful as a means to keep Dr. D. R. Hale (Cleveland, Ohio) (communicated) : 8 Sirota, J . Tech. Physics, U.S.S.R., 1948, 18, 1136. 9 Trommel, Naturwiss. (to be published). 10 This Discussion. 11 Hale, Science, 1948, 107. 393.364 CONCLUDING REMARKS the supersaturation from reaching high values when the more soluble vitreous silica is used. Skeletal, drusy or other irregular types of deposition on the seed imply the effect of a highly supersaturated condition, assuming that the solution is clean and not strongly agitated. A well-controlled crystal-growing system should preferably avoid all spontaneous nucleation, which is the recognized ideal in growing the usual types of easily soluble substances.

ISSN:0366-9033    

DOI:10.1039/DF9490500358

出版商:RSC    

年代:1949

数据来源: RSC

摘要:

364 CONCLUDING REMARKS V. CONCLUDING REMARKS Dr. C . W. Bum (I.C.I., Plastics) said: In looking a t the Discussion as a whole, and at the relations between theoretical and experimental contributions, certain things seem to me to stand out. In the first place, I am agreeably surprised a t the general agreement that a perfect crystal bounded by simple faces probably would not grow a t all. I suggested this as an inference from experimental evidence, but hardly expected i t to pass unchallenged ; actually it appears that there are theoretical grounds for expecting this to be the case. So far, theory and experiment are in accord. My other outstanding impression about the theoretical work is that much of i t is based on equilibrium considera- tions; yet crystal growth is, of course, not an equilibrium affair.I do not suggest that equilibrium considerations are irrelevant, but I think we should not assume that the conclusions from equilibrium considerations apply as they stand to actual crystal growth problems ; or, a t any rate, not to rapid growth. As they stand, they are likely to apply most closely to very slow growth. For the phenomena of rapid growth we need a dynamic theory, which treats crystal growth as a progressive event, takes into account the movements of the molecules in vapour, solution or melt, and considers how these movements influence the sites taken up on deposition and thus determine the character of the new surface on which further deposition is to take place. If we accept the thesis that perfect crystals bounded by low-index faces do not grow at any reasonable supersaturation, we are faced by the problem of accounting for the fact that crystals do actually grow, even at very low super- saturations.There are two obvious solutions; one is that real crystals are not perfect, the other is that the surfaces on which deposition occurs are not low-index surfaces ; and both these conceptions have figured in our discussions. No doubt both factors play a part in determining the rates a t which crystals grow, and we have to enquire what is their relative importance in a variety of circumstances. Turning to the experimental papers, they fall fairly sharply into two classes -those dealing with rapid growth and those dealing with extremely slow growth. The former are relevant to the industries which produce crystalline substances in large quantities and are necessarily concerned with rapid crystal growth ; the study of the rate of nucleus formation and the rate of crystal growth is directed towards maximum production and control of grain size and shape.The latter are the concern of the industries which make large perfect crystals for optical prisms or piezoelectric elements, and must necessarily grow their crystals very slowly in carefully controlled conditions. How are the industrial, experimental and theoretical aspects to be linked up ? It seems to me that i t is for problems of rapid growth from strongly super- saturated solution that a dynamic theory of crystal growth is most needed, and that since in rapid growth deposition apparently occurs on high-index surfaces a t the edges of spreading layers, the central problem is the study of the factors which keep high-index surfaces alive.On the other hand, in very slow growth from slightly supersaturated solutions there is time for high-index surfaces t o heal (that is, for the depositing molecules to go on to sites which give rise to low-index surfaces) ; in these circumstances, i t is likely that imperfections play a dominant role in controlling growth ; here, too, the theoretical approach based on equilibrium considerations is likely to be more directly applicable. Mr. P. R . Rowland (Londovt) said : The general view of the meeting seems to be that the gap between the theoretical and experimental approaches has been too wide. As a colleague To the author, the reason for this seems obvious.CONCLUDING REMARKS 365 remarked, “ The subject is still in the alchemical stage.” As an experimentalist the author feels that it is asking too much of the theoretical worker to provide even rough theories at this point of development.The information available is too meagre and the possible complications too many. However, we might have asked for guidance concerning the lines along which further work may be conducted. The author would therefore like to put forward a few opinions, though with some trepidation, since Prof. Stranski appears to have been following the course proposed and may already have forestalled them. First, the subject should be subdivided under the following headings : (i) Growth from vapour.(iii) Growth from solution. (iv) Growth by phase change in the solid state. Further subdivision according to whether the crystal is held together by ionic, homopolar, van der Waals’ or metallic forces also seems desirable. (ii) Growth from melt. Two aims should be borne in mind : (a) To provide a picture of the structures of the growing surfaces, meaning by “ picture ” the sort of information which is imparted by describing, say, methane or long conjugated chains in terms of cx bonds, etc. (b) A similar picture of the medium from which the crystal is growing, with due regard to the fact that the situation is dynamic and not static. The difficulty of supplying ( b ) will increase as we proceed from (i) above to (iii) ((iv) is a special case). In the case of growth from solution (a) and (b) may not be separable.For instance, when growing CuS0,.5H,O crystals there may be quite a large growing region, in passing through which Cu(H,O),++, SO4-- ions and water molecules gradually pass from a more or less random distribution to become units of a lattice correctly spaced and rotationally orientated. It would be difficult to say which point is the surface of the growing crystal. This picture is in fact supported by the observation that growth of good crystals from aqueous solution is only usually possible with hydrated substances, while growth from the melt seems to be the most successful way of producing really large perfect crystals. Can growth from solution be regarded as growth from a highly impure melt ? However, i t would seem wiser to start with the simplest systems (a) above.Crystallization may then be regarded as a heterogeneous reaction and it is essential that we learn as much as possible about the surface of the substrate. The author regards the technique of forming spherical single crystals and studying reactions on their surfaces as a powerful tool in the experimental study of solid surfaces. Prof. Stranski has used it to study crystal growth itself. It is suggested that progress from such beginnings may be made by carrying out work on the following lines : I . Experiments to determine the behaviour of the surface of single crystal spheres towards various reagents, e.g., the vapour of the crystal substance, adsorbates, substances of varying electronegativity, polar substances, solvents, etc.The only way to ensure a really clean surface is to heat i t in a vacuum and valuable information may be obtained by repeating some of J. K. Roberts’s work with single crystal wires. 2. Experiments with spheres a t temperatures very near their meltihg points to determine the mobility of surface layers. Growth from the vapour under these conditions, especially in cases where the gas phase could be made very dense, may give hints on the mechanisms of growth from the melt. 3 . Growth of solvated crystals from the melt may throw light on growth from s o h tion. 4. Dr. Bunn has shown how much is to be learnt by the direct observation of growing crystals. I n the electron microscope and interferometry as developed by Prof. Tolansky and his school, we now have methods which enable us to observe almost down to molecular dimensions. Though they have limitations, obvious and otherwise, the author is sure that if the attempt were made to adapt them to the study of growing crystals, a t least some confusion would be removed. The central problems could be recognized and attacked. To sum up, the theoretical physicist will only have a fair chance of getting to grips with the problem when the experimentalist has revealed what its essentials are, It is suggested that the best line of approach is to begin with a systematic study of crystal surfaces by both direct and indirect methods.

ISSN:0366-9033    

DOI:10.1039/DF9490500364

出版商:RSC    

年代:1949

数据来源: RSC

摘要:

AUTHOR INDEX * Amords, J. L., 75. Bannister, F. A., 291. Barrer, R. M., 326, 360, 362. Becker, R., 55. Bradley, R. S., 75. Bransom, S. H., 83, 96. Bruno, A. J., 112. Buckley, H. E., 243. Bunn, C. W., 75,119,132,193,287,364. Burton, W. K., 33, 40, 73, 184. Butchart, A., 254. Cabrera, N., 33, 40, 73, 184. Colegrave, E. B., 352. Correns, C. W., 267, 290. Convin, J, F., 172. Davies, C. W., 103. Denbigh, K. G., 188, 196. Dunning, W. J., 79, 83, 96, 183, 186, Egli, P. H., 61, 166. Ehrenberg, W., 358 Emmett, H., 119. Evans, U. R., 77. Fell, E. W., 360. Fordham, S., 117, 186, 187, 288, 289. Frank, F. C., 48, 66, 67, 72, 76, 186. 189, Franks, J. A., 358. Garner, W. E., 7, 187, 194, 362. Griffin, L. J., 192. Hale, D. R., 189, 363. Hall, E. O., 195, 196. Hardy, H. K., 69, 196. Hartshorne, N.H., 149. Haven, Y., 195. 358. Hocart, R., 237, 289. Holden, A. N., 312. Humphreys-Owen, S. P. F., 66, 144. Jaffe, H., 319. Johnson, L. R., 166. Jones, A. L., 103. Juliard, A , , 191, 285. 194, 195. 286. Kjellgren, B. R. F., 319. Kletz, T. A., 359. Lea, F. M., 289, 345. Lennard-Jones, J., 197. McCrone, W. C., 158. Mathieu-Sicaud, (Mlle.) A., 237. MenZies, A. C., 306. Michel-LBvy, A., 325. Millard, B., 83. Molibre, K., 21. hlonier, J. C., 289. Mott, N. F., 11. Nurse, R. W., 345, 363. Owen, G. E., 172. Plusjk, M. H. R. J., 183. Powers, H. E. E., 196, 285. Raistrick, B., 234, 286, 360. Rathje, W., 21. Rhodin, T. N., Jr., 215. Richardson, H. M., 352. Rigby, G. R., 352, 362. Robinson, A. E., 192, 315, 359, 360. Rowland, P. R., 284, 286, 359, 364. Seits, F., 271. Skinner, J., 306. Stockbarger, D. C., 294, 299. Stranski, I. N., 13,21, 69, 74, 75, 193. Strickland-Constable, R. F., 184. Swinnerton, A. C., 172. Thomas, L. A., 341. Ubbelohde, A. R., 180, 197. Van der Merwe, J. H., 201, 283. Van Hook, A., 112. Van Praagh, G., 338. Wells, A. F., 196, 197, 286. Whetstone, J., 254, 261, 289. Willems, J., 283. Woodward, P., 284. Wooster, Nora, 341. Wooster, W. A., 188,341. Wyart, J., 323. Zerfoss, S., 61, 166. * The references in heavy type indicate papers submitted for discussion 366

ISSN:0366-9033    

DOI:10.1039/DF9490500366

出版商:RSC    

年代:1949

数据来源: RSC