Kinetic study of the sulfurisation of cuprous halides in hydrogen sulfide

Mat. Res. Bull. Vol. I0, pp. 641-650, 1975. Pergamon Press, Inc. Printed in the United States.

KINETIC STUDY OF THE SULFURISATION OF CUPROUS HALIDES IN HYDROGEN SULFIDE

Jean-Pierre Larpin, Jean-Claude Colson Laboratoire de Recherches sur la R~activit~ des Solides, L.A. 23, Facult~

des Sciences Mirande, Dijon, France and Denise Delafosse

Laboratoire sur la Cin~tique des R~actions Superficielles, E.R. 133, Paris VI T 55, 4, Place Jussieu, Paris, France.

(Received February 18, 1975 and in final form May 7, 1975; Refereed)

ABSTRACT The reaction mechanism of CuCl sulfidation in H2S and in mixtures H2S/HCI is studied. The influence of the gas formed during reaction upon the values of the kinetic constants is shown experimentally on samples of different natures. The analysis of kinetic and morphological results leads to an interpretation taking into account, on the one hand, the prevailing step activation energy in order to determi- ne the transformation rate and on the other the adsorption heats of the r~active and formed gases, This work is only part of a more tho- rough study about cuprous halides sulfidation.

The kinetic aspects of solid-gas reactions in which one or more gases are formed are not as well understood as those of transformations in which the simple adsorption of gas atoms, or even decomposition, occurs. A few examples can be found in the literature such as the sulfurisation of metal halides (1-2-3-4). These studies demonstrate the importance of the gas or gases formed upon interfacial reactions. The precise nature of the in i t ia l solid is also of great importance, as we showed in a study of the sulfurisation of metal sulfates by hydrogen sulfide (5).

The role of solid or gaseous impurities in certain solid-gas reactions (6-7-8) has been determined. The availability of high purity monocrystalline cuprous halides encouraged us to continue and study in greater depth, previous work on cuprous halides. We report, in this f i rs t part, the results for chlo- ri de.

Preparation and characterization of the cuprous chlorides samples

Whereas dry air or oxygen have no effect, moist air leads to a very rapid oxidation of CuCl even at ambient temperature. The different methods used to prepare the samples studied in this work are given in Table 1.

A bar of polycrystalline CuCl, with purity better 99.999 % (9) was obtained by means of zone refining. I t is then possible to induce crystal

641

64Z SULFURISATION OF CUPROUS HALIDES Vol. I0, No~ 7

growth in this bar by a dry method and to obtain an optically pure monocrys- tal (I0).

TABLE I

Sample studied

Sample

Monocrystalline

Polycrystalline

Monocrystalline

Powder I

Powder II

Powder I l l

Preparation

CuCl > 99.999 % pure

II II

Recrystal I ization of CuCl in s i l ica gel

Crushed pieces of op- t ica l ly pure CuCl

Decomplexation of an HCI solution of CuCl

(i rreproducible results) Powder I I sublimed in He

Reaction conditions

T°C PH2 S (torr)

17.9 - 57.9

27 13 - 87

16 - 20 20

7 - 33 20 10.2 18 + 100

Results irreproducible

2 - 9 20

Samples prepared by other methods, such as have been previously studied, are contaminated superfically or even in bulk by oxychloride formation or HCI.

HCl and H2S were supplied by Mattheson and were carefully d is t i l led.

Sol.!.d phase obtained

Under standard conditions the partial pressures of hydrogen chloride and hydrogen sulfide necessary for the equilibrium 2CuCl + H2S • Cu2S + 2HCI are such that P~CI / PH2S = 10.6 At,

In the temperature and pressure ranges indicated (T1) the system tends, therefore, towards the formation of stoichiometric Cu2S of the ortho- rhombic variety. No other intermediate compound could be detected by X-ray spectroscopy. The values of the crystallographic parameters refine~ by the least-squares method are : a = 11.881 A b = 27,323 A c : 13,491 A,

The specific area increases considerably during transformation. I t changes from 0.5 to 8 m2.g-1 for example, when powder I undergoes complete reaction. The kinetic study was carried out in a thermobalance (5).

Morphological aspect of the transform.ation

Monocrystalline samples prepared by Nikitine and al. (9) were studied under a scanning electron microscope after sulfurisation for different pe- riods of time, such that the degree of transformation was between 10 and 100%. Attack develops over the entire surface of the sample and proceeds by formation of successive of ~Cu2S layers.

At low conversion ~ < 0.1 the external surface is completely covered with homogeneous chalcosine which separates easily from the substrate. At higher conversion the sulfide layer is more compact in the vicini ty of the chloride sulfide interface ; the outer layers contain numerous oriented chan-

Vol. i0, No. 7 SULFURISATION OF CUPROUS HALIDES 643

nels which emerge at the sulfide-gas interface (Fig. 1).

FIG. 1 FIG. Z

Cu2S layer detached from mono- Appearence of pores at the sur- crystalline CuCI face of Cu2S

The external surface of the sulfide retains the morphology of the original crystal ; i t contains certain macroscopic defects, in particular, regular fissures which must correspond to the emergence of pores (Fig. 2). The mode of attack on the massive polycrystalline sample is similar. The presence of inclusions and of macroscopic surface defects made the examina- tion of the monocrystalline tetrahedra obtained by recrystall isation from si- l ica gel impossible.

Our results appear to be in contradiction with those of (8) who observed enlargement of individual three-dimensional grains on the macroscopic defects of the monocrystals.

We believe, now, that this type of behaviour is related to crystalline imperfections in the samples studied or to certain superficial impurities. The results obtained from the kinetic studies, moreover, confirm this hypothe- sis.

Kinetic aspect of the transformation

Attack by pure H2S

Depending on the origin of the sample studied, there are marked di f - ferences in the form of ~ = f ( t ) (~ = mass loss at t divided by mass loss at reaction end) (Fig. 3) curves expressing the extent of transformation with time. The massive high purity mono or polycrystalline samples lead to ~ = f ( t ) curves corresponding to rapid and uniform germination over the whole in i t ia l surface, and the advance of a reaction front at constant rate ; in fact, the curves can be transformed into straight lines by the expression (1-~)1/3= f ( t ) (Fig. 3). The result implies that the massive pieces studied can be considered as spheres. This result is in agreement with the morphological observa- tions performed on the same samples. The curves ~ = f ( t ) are superimposable after an a f f in i ty ratio z in paralle to the t axis (~,t + ~,kt) up to

= 0.9, in the temperature and pressure ranges considered. The experimental

644 SULFURISATION OF CUPROUS HALIDES Vol. 10, No. 7

!

3 = 0 0 100 2()0 t mn 300

o,5

I

FIG. 3

= f(t) and (i - e)i/3 = kt curves for the

formation of chalcosite on a monocr~ystalli- ne CuCI sample with pure H2S

T : 33°C PH2 S : 20 tort

sured at ~ = 0.5 is E = 25 Kcal.M-l.

activation energy determined from instantaneous rates with identical ~ by plot- ting log v~ = f(1/T) is

E = 9.5 ± I Kcal.M "1, The overall rate of the transformation depends on the H2S pressure, and follows a la~ such as : n

v = kPH2 S with n ~ 0.5.

For the other types of massive samples and in particular for the tetrahe- dral monocrystals obtained from a sil ica gel, the results are variable : the ap- pearance of a pure induction period whose duration depends on the conditions of attack and the state of the sample has been observed. This induction period is followed by an acceleration period up to a conversion ra t ioo f 0.5. The value of the experimental activation energy mea-

The pulverised samples also give different results depending on the preparation and the size of the particles. The ~ = f ( t ) curves for a given superimposed show an induction period which varies with the transformation, the pressure and the type of pretreatment which the sample has been submitted to, Furthermore, the finer the particles the higher the conversion ratio at which the inflexion point occurs (Fig. 4 and Fig. 5).

The curves obtained with powder I (purity better than 99.999 %) are not superimposable after an aff ini ty (~,t ÷ ~,kt) up to ~ = 0,2 (Fig, 4!. The activation energy measured at ~ = 0,5 is 16 Kcal.M-i between 7 and 16% In all cases, however, beyond the inflexion point, the different sets of curves expressing the extent of transformation with time, can be converted into

t I , aws straight lines by the equation (i-~11/3 = f( Rate l with respect to pressure are the same as those obtalned for t e massive samples.

Attack by H2S/HCI mixtures

Those previous results suggested that the most important factor must be the reaction temperature which depends on the fineness of the in i t ia l so- l id and its overall reactivity.

These considerations led us to believe that the concentration of the gas phases present at the solid-gas interface had a bearing upon the advance of the reaction. We carried out a kinetic study of the transformation with HpS/HCl gas mixtures on powder-sample I either by varying the ratio of the p~rtial pressures (0.25 < PHcl/PH2s < 1) at a given temperature and total

Vol. 10, No. 7 SULFURISATION OF CUPROUS HALIDES 645

pressure (P = PH2 S + PHCl) or

by varying the temperature between 13.7°C and 480C for a given gas mixture (PHCl/PH2S=0.25)

at a fixed total pressure (20 torr) .

I t is observed that the presence of HCI in the gas phase slows the reaction below 40 ° but has no effect at higher temperatures (Fig. 6). Over the whole temperature range, the

= f ( t ) curves, beyond the point of inflexion curves can s t i l l be transformed into straight lines using the expression ( I -~) i /3 = f ( t ) ,

However, the curves are no longer superimposable an a f f in i ty (m,t ÷ m,kt). There are, in fact, two ranges, one for temperature above 39°C, the other below. Different values of the activation energy correspond to these two ranges. Finally, a constant temperature study with different pressures PT of a 20 % HCI gas mixture showed that the reaction rate varies l i t t l e but inversely as the total pressure (Fig. 7).

Discussion - Conclusion

The experimental results lead us to pay great at- tention to the overall mechanism to the surface concentra- tions of the reactive gas and that which is formed. The nature and the role of the pro- cesses occuring in the develop- ment of the systems can usual- ly be determined by decomposing

0 0 25 t mn 50

FIG. 4

e=f(t) curves for the formation of chalcosite on a pulverised CuCl sample with pure

H2S (Powder i).

/

I 0(.

0,5

0 125 t mn 250

FIG. 5

s=f(t) curves for the formation of chalcosite on sublimed CuCI sample with pure H2S

(Powder 2).

the reaction into elementary steps (11).

We shall consider the following steps :

a) Adsorption of HpS on the in i t ia l CuCl. el) Sulfur insertion into the lat t ice. e~) Desorption of HCI. e3) Formation of an epitaxic grain of CupS by supersaturation of the defects, f~llowed by reorganisation of the new phase into chalcosine. I f i t is assumed that step e3) is very fast and does not affect the overall transformation

646 SULFURISATION OF CUPROUS HALIDES Vol. i0, No. 7

rate i t is possible, by means of the steady state approximation, to obtain from the rate equations of steps al) el)ande2) an expression for the overall slowing down of the changing system per unit area.

KaKelKe2 IKelKe2(I+P~)+P~ KaKe2PI+Ke2+P2 PIKa(I+Kel)+I ]

~r = KaKelKe2PI-P ~ [ aaKelKe2 + eelKaKe2 + ae2KaKe2 (I)

Where aj = Kj.s o, ~j =~ .Soare the product of the number of superficial, sites

and the direct and reverse rate constants of steps j , Kj = aJ/~iv is the S O

pseudo-equilibrium constant of these steps. PI and Pp are the partial pressures of H2S and HCI respectively. The diffu- slon vacancies lead to the formation of pores and to separation of newly formed and reorganised blocks of sulfide. When the temperature is high enough this mechanism occurs uniformly over the whole surface since germination is rapid and supersaturation of defects is always attained.

I

. • , , ,

°i 0 10 2 t mn

FIG. 6

Effect of HCI on the reaction at different temperatures.

The size of the chalcosine blocks is limited by mechanical forces, and the blocks separate from the substrate to regenerate fresh chloride surfaces.

Macroscopically, we observe a reaction front made up of a mosaic of sulfide blocks separated from each other by cracks which allow access of the gas to the ini- tial phase.

Consequently, we can assume that the most important step is el) ; equation I then becomes :

+ 1 KaKelKe2 + Ke2

~= KaKelKe2P'I 'P~ L "aelKaKe2 (II)

1/ I f the partial pressure of HCI can be neglected, equation (I I) has the form :

SokelKaP1 V =

1 + KaP s

two limit cases can be considered : - KaP 1 << 1. The concentration of adsorbed species is low and the overall

Vol. I0, No. 7 SULFURISATION OF CUPROUS HALIDES 647

rate is V = sok~KaP 1. The activation energy will be Eexp = Ee I - QI" Where

E e the a c t i v a t i o n energy of s t ep e I and Q1 is the a d s o r p t i o n fieat of H2S on

th~ in i t ia l solid.

KaP 1 >> 1. The adsorbed l aye r i s s a t u r a t e d ; the ove ra l l r a t e V = Sokel

is independent of the gas pressure and the experimental activation energy will be Eex p = Eel . ki.cst

The second case does not correspond to our experimental ob- servations since the rate is never independent of the gas pressure.

The f i r s t case does not explain the observed variations in the .S experimental activation energy. The observed results cannot be interpre- ted by means of the simplifying as- sumption P2 = O.

2/ I f the partial pressure 4,~ of HCI cannot be neglected one ob- tains a complex expression for the overall rate which can be simpli- fied i f the following two assumptions are made, - The adsorbed layer is saturated. - The rate of reinsertion of chlo- 4 ride in the neutral defect created is very small :

cst

k i

PHCI - 0.25

PH2S

T :26.8 C 0,3

SokelP~ ÷

<< SoKaPlkel Ke 2

Equation II then becomes :

V =

P2 and i f r = - -

P1

I 0 ,2 0 .SO 100 Pt torr

FIG. 7

20 % HCI gaseous mixture total pressure effect upon the reaction rate at constant temperature. Total pressures

PH2S+ PHCI = 20 torr

Soke I

' +

Sokel V = ( I l l )

I + --!-1 r2 KaKe2 ~ PT

This expression is similar to that used in the kinetic heterogeneous catalysis when the reaction occurs in the presence of one or several extra- neous inhibitions. With this equation the experimental results obtained for

648 SULFURISATION OF CUPROUS HALIDES Vol. I0, No. 7

high purity powder samples exposed to H2S/HCI mixtures can be understood. In- deed, according to equation I l l the overall rate is inversely proportional to the total pressure and to the value of the ratio r as observed. I t is moreover, possible to explain the dependence of the experimental activation energy upon the temperature range :

1 - When r is so small that i << 1 Eex p = Eel

KaKe2P~/P1 which is the case of very HpS rich mixtures and the l imi t is reached when HCl is introduced into the gas ~hase only by the sulfurisation reaction.

2 - When r is no longer negligible and corresponds to the experimental 1 P~

values (0.25 < r < I) the term Ka-~l~e2-1~ 1 is s t i l l negligible when KaKe2iS

large that is, at suff ic ient ly high temperatures. Then Eex p = Eel, which is what is observed at temperatures above 40 ° for an HCI/H2S ~ixtur~.

2Conversely at lower temperatures KaKe2 is not great enough to make

1 P2 negligible.

The apparent activation energy no longer follows Arrhenius' equation and depends on the magnitude of Eel, Q1 and Q2 where Q2 is energy of adsorption of HCI on the in i t ia l solid C~CI.

1 r 2 At the l imi t i f r and KaKe 2 are such that ~ - ~ 2 ~ P T >> 1, the

experimental activation energy has form IV :

: + 2 Q2 - Q1 ( I V ) Eexp Ee 1 Above 40°C we observe a value of 15 Kcal/M -I in agreement with equa-

tion IV. In the temperature range considered here i t can be taken that the l imi t values of Q1 and Q2 correspond to a physical adsorption and are such that : 3 < QI < 8 Kcal.M-i and 2 < Q2 < 7 Kcal.M - I ,

We determined by open system isothermal microcalorimetrie (12)HQ2cI value being 7.5 ± 0.5 Kcal.M-1, which confirms our assumptions. The concentration at the interface cannot be neglected when the attack is effected by the pure gas. However, the above relationships do not provide an adequate basis for the interpretation of the results obtained on less pure samples and in part icular, the very high activation energies,

In al l cases, the reaction temperatures are very low because of the high react iv i ty of the samples ; i t is, therefore, possible that, apart from HCI concentration at the chloride-sulfide interface, step e3) (related to the reorganisation of the chalcosine lat t ice) is also rate l imit ing. Those results reveal the importance of the part played by the solid or gas impurities in the sulfurisation of CuCI. Only with a high purity sample is i t possible to interpret the kinetic data sat is factor i ly by breaking the reaction down into elementary steps and relating them to the morphological observa- tions. Under these conditions the overall transformation rate is limited by the interfacial reaction and depends on the partial pressures of HCI and H2S at the reaction interface. The experimental activation energy is a func- tion of the activation energy of the interfacial step i t s e l f but also the

Vol. 10, No. 7 SULFURISATION OF CUPROUS HALIDES 649

heats of adsorption of the gases present on CuCl depending on the temperature range of the transformation.

References

1. R. Mole, Thesis, Paris (1952).

2. M. Cadoret, R. Aubry and S. Vignal, C.R. Acad. Sci. Paris 274, C, 1013

(1972).

3. A. Pelloux, Thesis, Grenoble (1969).

4. J.C. Colson, D. Delafosse, P. Barret, C.R. Acad. Sci. Paris 258, 2825

(1964).

5. J.P. Larpin, D. Delafosse, Bull. Soc. Chim. Fr. 8, 3016 (1972).

6. B. Delmon, R~union Annuelle de la Soci6t~ de Chimie Physique, Dijon, Ju i l le t (1974), Elsevier Publ., Amsterdam (to be issued).

7. V. Boldyrev, 25~me R~union Annuelle de la Soci~t~ de Chimie Physique, Dijon, Ju i l le t (1974), Elsevier Publ., Amsterdam (to be issued).

8. G. Jeannin, D. Delafosse, P. Barret, C.R. Acad. Sci. Paris 265, 213 (1967).

9. C. Schwab, R. Heimburger, J. Ringeissen and S. Nikit ine, Brevet C.N.R.S. n ° 100-138 (1967).

10. S. Minoru, I. Rikizo, K. Yoschihiro and O. Takahiro, J. Electrochem. Soc. 114, 388 (1967).

11. P. Barret, Cin~tique H~t~rog~ne, ed. Gauthier-Villars, Paris (1973).

12. C. Letoquart, F. Rouquerol and J. Rouquerol, J. Chim. Phys. 3_, 559 (1973).

Documents

Kinetic study of the sulfurisation of cuprous halides in hydrogen sulfide