SHEAR TRANSFORMATION IN SOLID 1 → SOLID 2 + GAS ENDOTHERMIC DECOMPOSITIONS

(1)

HAL Id: jpa-00217275

https://hal.archives-ouvertes.fr/jpa-00217275

Submitted on 1 Jan 1977

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

SHEAR TRANSFORMATION IN SOLID 1

→ SOLID 2

+ GAS ENDOTHERMIC DECOMPOSITIONS

J. Niepce, G. Watelle

To cite this version:

J. Niepce, G. Watelle. SHEAR TRANSFORMATION IN SOLID 1 → SOLID 2 + GAS

(2)

JOURNAL DE PHYSIQUE Colloque C7, supplhment nu no 12. Tome 3 8 , dkcembre 1977, pnge C7-365

SHEAR TRANSFORMATION

IN

SOLID

1

4

SOLID 2

+

GA$

ENDOTHERMIC DECOMPOSITIONS

J. C. NIEPCE and G. WATELLE

Laboratoire de Recherches sur la RCactivitC des Solides (*). FacultC Sciences Mirande, F 21000 Dijon, France

RkumC. - Contrairement B ce qu'avaient fais& penser les Ctudes cinktiques il est montre que le passage de l'edifice cristallin d'un hydroxyde de type brucite en oxyde semble Etre une transformation non destructive c'est-a-dire que I'etat ordonne triplement pkriodique du solide initial semble Btre conservk dans cette transformation. Elle est topotactique et prksente un plan invariant. Les dimen- sions des cristallites produits en dehors des zones de frittage sont indkpendantes du degrk d'avan- cement de la reaction, de la tempkrature de dkcomposition et donc de la vitesse globale de celle-ci.

Abstract.

-

On the contrary of kinetic studies, the present results show that the structural transformation which occurs between a brucite type hydroxyde and the correspondent oxide, seems to

be a non-destructive transformation i.e. the periodic piling of the initial solid is preserved. The transformation is topotactic and exhibits an invariant plane. Outside of sintering regions, the obtained crystallite sizes are independent of fraction reacted of decomposition temperature and of course of the overall reaction rate.

Introduction.

-

Many endothermic decomposition

reactions of a crystalline solid into another, with gas formation, have been described by models using nucleation and growth concepts. So an intermediate disordered state is implicitely assumed and these transformations are often called reconstructive trans- formations. Indeed, some of them are known as topo-

tactic transformations; on the contrary, this term involves the idea of the preservation of an ordered state.

Thus, for such reactions, two contradictory concepts are used simultaneously. We have tried to analyse this contradiction in the case of the widely studied decompositions of brucite type hydroxides.

Experimental. - In the present study, cadmium hydroxide was used preferably to other isotypic hydroxides of brucite owing to its low decomposition temperature. Any risk of sintering, which always occurs when the oxide preparation temperature is higher, was thus obviated. The decomposition experimental conditions [l. 21 and the crystallite size calculations [3] have- been previously described.

Results : structural transformation features.

-

STRUCTURAL FEATURES. - The magnesium hydroxide decomposition reaction is one of the oldest topotactic reaction known [4]. That of calcium [5, 61, cobalt

[q

and cadmium [8] hydroxide is also topo-

tactic and exhibits the same orientation relationships. These relations

-

R - are listed in figure 1 which is an electron diffraction pattern obtained during the decomposition of a cadmium hydroxide crystal [9]

[OOl], -+ [l 1 11, H = hydroxide (1 120)" -, (220), 0 = oxide

.

The structural relationship between both com- pounds shown in figures 2 and 3 easily explain the

2110

0

.022

-

1100

olio

.

l 1120

FIG. 1. - Diagrammatic electron diffraction pattern of Cd(OH)2 crvstal during decomposition into CdO. Electron beam parallel to

[OOI],. spots of Cd(OH), ; Ok, spots of CdO with

(*) AssociC au C.N.R.S.

(3)

J. C. NIEPCE AND G. WATELLE

FIG. 2. - - Sct of two hexagonal anion layers enclosing a cation

layer : a) in the hydroxide structure. Projection on (OO.l),; h) in the ox~de structure. Projection on ( l I I),.

FIG. 3. -- Cross-sections of the three structures along the likcly shear-plane : a ) Cd(OH), a l o n g ( l 1 ~ 0 ) ~ ; b) Likely lacunar hydroxide

structure along (1 l?O)H: c) CdO along (220),. AI1 the presented atoms are in the shear-plane.

orientation relationships when Goodman's homo- geneous mechanism is assumed [10]. The water mole- cules form between two adjacent layers of hydroxyl groups and escape by a diffusion process. Then, the structure collapses along the C-axis of the hydroxide associated with a translation t , or t , of some of the

preserved ion layers as shown in figures 3a and 3c

DYNAMIC

FEATURES.

-

The following results were

obtained from a study of the division state of the oxide produced by means of X-ray diffraction peak profile analysis. Nevertheless, such a study can only be valuable if no evolution of the crystallite orga- nisation such as sintering or solid-gas interaction occurs after or during crystallite formation. So, preliminary work proved necessary. The influence of experimental decomposition conditions on oxide crystallite size was studied to determine the tempe-

rature range in which the oxide crystallite size is only related to the structural transformation mode [l, 21.

Figure 4 and table I show that, in vacuum, oxide crystallite size is independent of decomposition temperature. Above 300 O C , size increase is due to sintering

but the formation size is always independent of decomposition temperature. Now, the overall decomposition reaction rate is closely related to temperature. So, whatever the overall reaction rate the sizes of the oxide crystallites produced are the same.

FIG. 4. - Mean crystallite sizes D,, of oxides obtained at various temperatures in vacuum (decomposition time : 18 hrs). from

l l l peak ; 0 from 200 peak ;

+

from 220 peak.

Crystallite sizes for CdO obtained by Cd(OH),

decomposition at various temperatures, in vacuum for 6 hours. Determinations were made from 220 dijl fraction peak

D, by half peak breadth ; E @ , by integral breadth ; zIF, by integral breadth calculated by cosine coefficients of the Fourier series;

E,, by cosine coefficients of the Fourier series; Q, by slope of the variance function.

In these experimental conditions, another impor- tant result was obtained : oxide crystallite size is independent of the fraction reacted-a (table 11). sr is defined by ratio AnlAm, in which Am is the weight loss recorded and Am, the weight loss at the end of the reaction.

Mean apparent crystallite size D versus fraction reacted cc (vacuum and T = 164 OC). D has been

calculated from the h a u width of 111 oxide X-ray

(4)

SHEAR TRANSFORMATION IN SOLIDS ENDOTHERMIC DECOMPOSITIONS C7-?h7

The experimental results reported were obtained by studying cadmium hydroxide decomposition ;

magnesium hydroxide yields similar results ; however, owing to a higher decomposition temperature, the temperature range in which these phenomena can be observed in narrower [2].

Discussion and further results. - Many topotactic

transformation have been explained by comparing the initial and final structures but the interpretation proposed for orientation relationships can only be satisfactory if the crystalline edifice is not destroyed during transformation [Ill. Hence, such an interpretation implies that the transformation would be ordered. But orientation relationship are no evidence of that ordered transformation ; the only thing which is certain is that the interpretation proposed is only valid if the transformation is ordered.

The results obtained by crystallite size deter- mination d o not show a growth step of the new phase as predicted from kinetic studies [12, 13, 141. Even more, the fact that the oxide crystallite size is independent of the hydroxide decomposition temperature and therefore of the overall reaction rate excludes almost certainly that the new phase i.e. the oxide, is generated through a diffusion process of these constituants.

Thus, the structural transformation associated with this decomposition reaction exhibits the features of a diffusionless tra'nsformation for which the piling order conservation is the main characteristic [15].

Two other results make this conclusion more valid :

- The oxide crystallite shape is accounted for by a non-isometric fragmentation of the initial solid governed by strenghs involved by OH vacancies and then by the structural transformation [16, 171. -- An invariant plane family (11%)" and the evivalent families (1210), and

(21

10), which become (220),, ( 2 0 2 ) ~ and ( 0 2 3 ) ~ possibly act as a semi- coherent interface between both solids during the structural change. This has been put in evidence more accurately thanks to the presence of a trans-

[ I ) EA. N and NII.IY I:. J C . . Sit;< c i i c ~ liid. 10 ( 1977) 413.

121 DUMAS. P., EA. N., NILPCE. J . C. and WATELLE, G., J . Solid.

State Chem., to be published.

(31 NIEPCE, J. C., WATELLE, G. and BRETT, N. H., J.C.S. Faraday I, to be published.

(41 GARRIDO, J., Amer. Min. 36 (1951) 773.

[S] CHATTERJI, S. and JEFFERY, J. W., Min. Mug. 35 (1966) 867. [6] BRETT, N. H., Min. Mug. 37 (1969) 244.

[7] FIGLARZ, M. and VINCENT, F., C . R . Hebd. Skan. Acad. Sri.

266C (1968) 376.

[8] NIEPCE, J. C., WATELLE, G. and CLINARD, C., C.R. Hebd.

Skan. Acad. Sci. 274C (1972) 96.

191 NIEPCE, J. C.. Thesis, University Dijon (1976).

formation twin [l81 for the quite similar transformation between CdCO, and CdO.

Conclusion. - The analysis of these experimental

results shows that the structural transformation associated with the decomposition reaction of brucite- type hydroxides into oxide may be related to a shear transformation that is a non-destructive transformation. The results of kinetic studies did not allow such a conclusion. This is not oversurprising ; indeed, it is almost absolutely proved that interface flows, due to gas removal, have a prevailing influence on the rate of such a reaction [19]. A kinetic study, usually, gives little information about the structural transformation mode. A rather typical example illustrating this is given by the polymorphic transformation anatase-rutile in which no problems of matter transfer occur. This transformation has been explained by a nucleation-growth process from a kinetic study [20], and by the same authors, as a topotactic transformation (211.

How then can we consider the whole decomposition reaction ? Proposals have already been discussed [9, 221. Briefly, it may be argued that, first, water mole- cules are removed from the crystalline edifice of the hydroxide, which is left in a transitory lacunar state as illustrated in figure 36 ; cations would occupy all the octahedral sites of the hexagonal piling of oxygen ions. The structural transformation would then occur between this ordered lacunar phase and the stable phase i.e. the oxide. Such a sequence has effectively been observed during the dehydration of barium acid oxalate [23]. It is thus clear that the structural transformation occurs at constant composition ; hence it can be related to a shear transformation since no atomic displacement greater than a cell fraction is required.

In this example, therefore, the distinction between topotaxy and endotaxy [24, 251 - i.e. solid state transformations whose product has a three-dimen- sional orientation compared to the precursor and during which atoms are gained or lost in the former and a constant composition required in the latter -

does not seem to be basically justified.

110) GOODMAN. J. F.. Proc. R . Soc. London A 247 (1958) 346.

[ I I] BUEKGER. M. J., Fortsch. Minerulog. 39 (1961) 9.

[ l 21 GREGG, S. J. and RAZOUK, R. I., J. Chem. Soc. 1 (1949) 536. [I31 PAVLJUCHENKO, NOVOSELOVA and LIBRANT, P'estti. Akad.

Nauk. U.S.S.R. Khim. Navuk. l(1969) 6.

[l41 ANDERSON, P. J., HORLOCK, R. F. and AVERY, R. G., Proc'

Brit. Ceran?. Soc. 3 (1 965) 33.

[IS] BURKE, J., The kinerics of phase transformations in metals

(Pergarnon Press Publ.-Masson edit. Paris) 1968, 211. [l61 NIEPCE, J. C., MESNIER. M. T. and L ~ U E R , D., J. Solid. Store

Chent. 22 (1 977) 341.

(5)

C7-368 J . C. NIEPCE A N D G . WATELLE

[l81 FL.WUFT, N . and NIEPCE, J. C., J . Mat. Sci. 12 (1977) 1052; (221 NIEPCE, J. C. and WATELLE, G., C.R. Hebd. S a n . Acad. Sci. J Mtrr. Sri. 13 (1978). 276C (1973) 627.

[l91 BEKTRAND, G . , LALLEMANT, M., MOKHLISSE, A. and WATELLE, [231 MUTIN. J. C. and WATELLE, G., J. Solid. State Chem., to be

G.. J. Inorg. Nucl. Chem., to be published. published.

[24] SCHNEIDER, H. G. and RUTH, V., Advances in Epifaxy and (201 SHANNON, R. D. and PASK, J. A., J. Amer. Cernm. Soc. 48

(1965) 391. Endotaxy(VEB. Deutscher Verlagfiir Grundstofindustrie,

Leipzig) 197 1 .