HAL Id: jpa-00219777
https://hal.archives-ouvertes.fr/jpa-00219777
Submitted on 1 Jan 1980
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.
CHARGE TRANSFER AND ATOMIC VOLUME EFFECT IN FexSn1-x AMORPHOUS ALLOYS
M. Piecuch, Chr. Janot, G. Marchal
To cite this version:
M. Piecuch, Chr. Janot, G. Marchal. CHARGE TRANSFER AND ATOMIC VOLUME EFFECT IN FexSn1-x AMORPHOUS ALLOYS. Journal de Physique Colloques, 1980, 41 (C1), pp.C1-251-C1-252.
�10.1051/jphyscol:1980184�. �jpa-00219777�
JOURNAL DE PHYSIQUE Colloque Cl , supplkment au n
O1, Tome 41, janvier 1980, page C1-251
CHARGE TRANSFER AlrD ATOMIC VOLUME EFFECT I N F+nl, MZORPHOUS ALLOYS
M. Piecuch, Chr. Janot and G . Marchal
Laboratoire de Physique du SoZide (LA n o 155) - FacuZte' des Sc7:ences, CO n0140 - 54037 Nmcy Cedex fFrance).
I-- INTRODUCTION. phous alloys /13/. Thus in alloys of the iron-rich
Amorphous alloys have been obtained over a side the electronic density is larger near iron and wide compositional range and thus have provided a lower near tin atoms than in the alloys of the good opportunity of studying in details electronic
Otherside'
properties of alloys. The old controversy of the charge transfer has been reopen recently in attempts to understand the magnetic behaviour of amorphous alloys made of transition metals MT and metalloids Me (MT2el-x). Obviously, magnetization measurements /I/, in particular the existence of a critical value Xcr= 0.4 for the appearance of magnetic interac-
tions /2/ are in favour of a progressive filling of the 3d states of MT by transfer from the metalloid a-like band. However, band structure calculations in the Fe3Si crystalline compound /3/ have shown that the sp state energy is lower than that of the Fe-3d states, making impossible any charge transfer /4/.
On the other hand the critical composition can be interpreted in terms of several different models /2/
including a localized model / 5 / which is an exten- sion of that once proposed by Jaccarino and Walker 161.
The FexSnl-x amorphous alloys is a good can- didate to make valuable contribution to this charge transfer problem since electron density measurements can be carried out simultaneously on 5 7 ~ e and 'l9sn nuclei using Mijssbauer spectroscopy. Details about
preparation, electrical resistivity measurements,
Fig. 1.
Composition dependence of the mean iso- mer shift measured on 5 7 ~ e (a) and
l19sn (b) (B). in Fe Snl,x amorphous alloys.
Full lines : calculated values ;
XCuSn amorphous alloys /13/ ;
V crystalline iron-tin compounds /12/.
111 - INTERPRETATION AND DISCUSSION.
Changes in electronic density may arise fro&
atomic volume effect and charge transfer effect.
Thus this section will try to analyse the relative density data, magnetization values and Miifdssbauer ex- contribution of both ef
According to Watson et al. /14/, a mean ato- periments in these amorphous alloys have been exten-
mic volume in binary amorphous alloys is given by sively reported elsewhere /2/ /7/ /8/ /9/ /lo/ and
the Pxpression : this present paper will concentrate on the interpre-
tation of isomer shift data.
11 - EXPERIMENTAL RESULTS.
Efdssbauet spectra measured at RT on 'l9sn and 5 7 ~ e isotopes can be mainly described in terms of broad, more or less asymetrical sextuplets for PexSnl-x alloys in their magnetic form. As explai- ned elsewhere /11/ a composition dependent mean is isomer shift can be defined for both tin and iron and determined from the geometrical centroid of the spectra. These mean isomer shift data are presented in figure 1 along with results obtained from crys- talline iron-tin compounds /12/ and CuxSnll amor-
with <M(x)> = x M(Fe) + (I-x)M(Sn) in which M(F~) and M(s~) are the molecular weight of iron and tin ;
#is the Avogadro number and d(x) the density values as measured in ref. /7/. The experimental data shows a linear composition dependence of <W (x)> in
at
PexSnl-x amorphous alloys. This is an experimental evidence that the mean atomic volumes of both Sn and Fe are mostly constant whatever the alloy composition 1Iowever, it is worth noting that this constancy may result from local fluctuation averaging to zero in space. Fitting the data with the formula :
18
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1980184
C1-252 JOURNAL DE PHYSIQUE
<2fat(x)> = x%Fe) + (1-x) % ~ n ) leads to : Ga,(~e) 15.7 i3
% d s n ) = 34.6 i3
to be compared with 'LT (Fe) = 11.7 i3 in the bcc crys- at
0
3
talline structure and 'Vat(Sn) = 27 A in the tetra- gonal structure of e t i n . Thus changes in isomer shift values have to be ascribed to charge transfer only. The genespi krmula for isomer shift can be
written :
6 s 6 +ctAp(o)
in which 6, is the isomer shift in a reference ma- terial and ?(o) is the electronic density at the nucleus.
In iron compounds, P(o) includes contribu- tions from core electrons Pc(o) and conduction band electrons Pb(o). According to Ingalls et al. 1151
in bcc a-Fe. In a close packed structure like &-Fe or amorphous alloys, ec(o) i$ still the same but
-- -
fb(o) should be slightly larger. However there is an opposite variation in.amoiphous alloys coming from the atomic volume increasing from 11.7 1 3 (a-Fe) to 15.7 i3. As a reasonable estimate Pb(o) will be ta- ken around 7 a : n 4 , Thus isomer shift of iron amor-
. -
phous compound may be given by the expression :
& ~ e y) = 60 -a [7 A %s - 5 ~ n ) ~ ] ( 2 ) in which d is a positive constant of the order of
3 -1
0.23aomm.sec 1161.
In tin compounds there is still some contro- versy about the oL coefficient (Eq. l) which is found around 0.05 a mm.sec-' 3 when obtained from calcula-
I
ted relativistic wave functions or near 0.1 a ' arm.
sec-' if deduced from electron conversion expeQriments /17/. In this paper fsn(o) has been calculated using the ue Vries et al. I181 wave functions and conse-
3 -1
quently cLZ0.05 a mm.sec , thus leading to the expression :
d(sn
X )= Jto + 1 . 0 9 A n ~ ~ ( 3 ) The electronic density at 'l9sn nucleus is larger in amorphous than in crystalline material, despite a bigger atomic size ; it must be ascribed to reducing in covalency effect. Thus we will assume that an ideal amorphous pure tin metal would have a
2 2
5s 5p configuration and then Q0(a-sn) 2 3.27 m.
see-' (with respect to BaSn03). In thb same way, it can be assumed that an ideal amorphous pure iron w t a l would have the same electronic configuration
than metallicd-Fe, that is something like 3d7'54s0.5, the difference in Qb(o) leading to m isomer shift of = + 0.14 m . s e c l for pure amorphous iron with ,respect to cipstallined-Fe.
Calculation of charge transfer are made here through a simple phenomenological approach wi- thin the following hypotheses :
- the Sn 46 states, being about 30 eV below the Fe 3d states, are completely filled ;
- the electronic density in the conduction band has a constant homogeneous value inside the
."
alloy ; thus lVat(Sn) * 2 Gat(Fe) results in
nb(sn) 2 2 %(Pel (4)
- near the iron atoms the conduction electrons are mostly 4s like, that is %(Fe) E nls(Fe) (5)
- near tin atoms the conduction electrons are half 5s, half 5s-like, that is nb(sn)*2 nSstsn) (6)
- the average number of electrons transferred from sp tin band to 3d iron state is in proportion of tin concentration and the 2.5 holes in the 3d iron state are full up when an iron atom is surroun- ded by tin atoms only, so : A n (Fe) 2.5 (1-X) (7)
3d
Equations (41, (5) and (6) results in n (Sn) = 5 s n4s(~e) which gives : An5JSn) - 1.5 +dn4s(Fe) ( 8 )
Using the expressions (7) and (8) in the elec- tronic balance of the material :
leads to
( F ~ ) 3 - 5.5 x + 2.5 x 2
n4s 2 - x
Putting these values in Eq.(2) and (3) along with hnjd(Fe) (Eq.7) gives :
The agreement with experimental data is quite reaso- -
nable in figure 1
nsrcP.eKcEs
I l l O'BUsDLEY R.C. and EQUDwm: D.S., Phys. Stat. Sol. (a)
9
- (1978) bQ7.I21 nAUGIR Ph.. P I e N C B I., nW.CiU C. and JANOI Chr., 3. of Phy.ics
r.:
Iht.1 Phy8iss ) (1978) 2083.131 S3tTmiDICK A.C.. Sol. Stat.
-. 2
(1976) 111.I41 WE3 P., BUDNICX 3 . 1 . md C M C I U C.S.. i n "U.tallie C1.ss.s". Ed.
bt
Leamy and Oil- 3 . 3 . r\D( m e a l s P a d , Ohio. 1977. Q.P. 11.131 TURBAN L . e t W M I U P . . 3 . Php. C. Solid Stat* Physic.. v o l . 12 (1979) 961.
161 . l A C w I B O V. and WILKZP L.R., Phya. I n . t.ttar*,
12
(1963) 238.171 (a) M(LeRU C.. MBCXU Ph. PfrCDCd 8 . . ROmvSQ 8 . ard J m T Chr. ?roc.of ehe 3rd fnt. C o d . oa r a p i i l y quenched m8t.l. (Brifiton)
2
(1978)'73.(b) WJZCXtI K. md tUB3ML C . , Phys. Letters
686
(1978) 466.181 IWLcmL
o.,
W C X a Ph.. PrrCDcsn.
md J A O I hr.. m t . Scisnc. Emg. (1978) 11.191 HOD~UCQ 1.. PICCVCB
n.,
~ARCW c . , ~UNCIRn.
md J ~ O I hr..m
T ~ S ~ S . -8.(1978). a' 2, 001.
1101 POD~UCQ 8 . . PIECUF
n.,
JAWT m r . . M B C ~ ~ L c. a d IV.I(OINm.,
~ h y * . ~ r r . (co ba t&lish8d).1111
W a n
Ph., IWLcmL C.. P I l C V c s )I. md JAno'I h r . , 3. of Phys. E2
(1976) 1101.1121 TXW27 C., BOZH I.. DIECA-PARIABASSOU C. d UcOep P., Phy8. 8.v. B
2
(1977) 3477.I131 B o W J. and P O M U I.. Zeit. Kp.. (1975) 95.
/I&/ VATSOW P.C. md BEBHdl L.U.. in "Char* Tim.f.r. L1.atronie Struetmrs o f .Ilo,s"
l h t . Joe. o f A I h l (1973). chap. 1.
l l S l I L L S P., VN8 DFS Hxmr
s.
ard S A W A ' C.A.. i n 'Ws.b.ur 1r-r Shift".Llorth Relland (1978) 361.
I161 D m K.J.. P h p . Bn.
19
(197&) 66.1171 RIUU P.A.. in "KUssbaur t o r r Shift". liortb Aol1.d (1978) 593.
1181 DS V U n S J.L.K.P.. TXUJSl3R J.M. md POS P.. 3 . Cha. Phys.
