• Nem Talált Eredményt

TI С 38.

N/A
N/A
Protected

Academic year: 2022

Ossza meg "TI С 38."

Copied!
16
0
0

Teljes szövegt

(1)

TI С 38.

А С

KFKI-71-43

nyvtára

AI • дат

I. Vincze

AVERAGE MAGNETIZATION O F Fe-AI ALLOYS

S ^a ri^axinn S^cademy o f (Sciences

CENTRAL RESEARCH

INSTITUTE FOR PHYSICS

BUDAPEST

(2)

ю

(3)

KFKI-71“43

A V E R A G E M A G N E T I Z A T I O N OF F

e

~AL A L L O Y S

Vincze

Central Research Institute for Physics, Budapest(Hungary Solid State Physics Department

A shortened version submitted to Physicei Status Solidi

(4)

A B S T R A C T

It is shown that the average magnetization of Fe-Al alloys can be well described if the iron moment is assumed to change markedly with a give number of Al first neighbours of the Fe atom. For 4 Al first neighbours the Fe moment was determined from the reported magnetization data as

yA = 1.82 + 0.02 Ув» To account for the magnetic behaviour observed above concentrations of 25 at% Al the presence of magnetic Fe^Al type clusters is assumed.

РЕЗЮМЕ

Показывается, что средняя намагниченность сплавов

fe-Ai

может хорошо описываться, если предполагается, что магнитный момент атомов железа является характерной функцией числа первых соседних атомов алю­

миния. На основе сообщенных данных по намагниченности был определен момент атомов железа в случае четырех первых соседних атомов алюминия:

уА = 1,82+0.02

v b

* Д ля описания магнитных свойств, наблюдаемых при кон­

центрациях выше 2 5 ат.% алюминия предполагается существование групп типа

Fe3Al*

KIVONAT

Megmutatjuk, hogy a Fe-Al ötvözetek átlagmágnesezettsége jól leír­

ható, ha feltesszük, hogy a vasatomok mágneses momentuma az első szomszéd Al atomok számának jellegzetes függvénye. A publikált mágnesezettség

adatokból meghatároztuk a vasatomok momentumát 4 Al első szomszéd esetén, melyre Уд = 1.82 + 0 . 0 2 yß adódott. A 25 at% Al feletti koncentrációknál megfigyelt mágneses viselkedés leirására mágneses Fe~Al tipus clusterek lé­

tezését tettük fel.

(5)

I N T R O D U C T I O N

Fe-Al alloys have been observed to exhibit up to 54 at% Al con­

centrations three different structures of their b.c.c. lattice, namely, disorder from О to '18 a't% Al, Fe^Al /or D03/-type order from 18 to 38 at%

Al and FeAl /or B2/ -type order above 38 at% Al. The b.c.c. lattice of the alloy comprises four f.c.c. sublattices, the sites of which are oc­

cupied by the different atoms as shown in Fig. 1 for Fe^Al and FeAl order.

Fig. 2 shows the average magnetization data, as extrapolated to T = 0 °K [l,2,3,4*1. For the explanation of the anomalous decrease in the average magnetization at about 30 at% Al observed by susceptibility meas­

urement, the presence of an antiferromagnetic FeAl phase [l] or the pre­

sence of a finely dispersed ferromagnetic Fe^Al phase with superparamag- netic behaviour in this antiferromagnetic FeAl phase [.4,5] have been sug­

gested. However, the diffuse X-ray scattering data /Fig. 3/ [б] and Mössbauer spectra [5,7,8] indicate a homogeneous transition. Neither the coexistence of two phases [5] nor the existence of a long range antiferro­

magnetic order in the FeAl phase [9] could be confirmed by experiment.

In contrast with the average magnetization [ 1,4 J and neutron dif­

fraction [9] measurements which do not show the existence of any long range magnetic order in the alloys with about 50 at% A l , it 1з apparent from Mössbauer spectroscopy [5,7] that only the Fe atoms at B- and D-type sites have a magnetic moment which can produce a short range order. This nearly ferromagnetic behaviour is confirmed also by the superparamagnetic specific heat anómaly observed at 48,8 at% Al [lO]. If the concentration of iron atoms increases, their number at sites A with 4 iron neighbours at sites D statistically also increases and Fe^Al type clusters can form which have sufficiently high moment to initiate a macroscopic magnetic ordering. However, at this stage the interaction between the clusters is still weak and that explains the observed anomalies of superparamagnetic nature [4 ].

The aim of the present work is to show that the changes in the average magnetization measured on Fe-Al alloys up to 50 at% Al can be ac­

counted for by taking into consideration the changes in the occupation of the sublattice sites by using Beck's assumption [7, ll].

(6)

2

M E T H O D OF C A L C U L A T I O N

The relation between the magnetic moment of the Fe atoms and the number i of first neighbour Al atoms is approximated as

f

yt) i = О, 1, 2, 3 VA if i = 4

О" i = 5, 6, 7, 8 /I/

where yD = 2.21 yß is the moment of pure Fe, while Уд was evaluated from the reported average magnetization data [l, 2, 3, 4] by the least- -square method as y. = 1.82 + 0,02 У_.

A В

We take рд , pß , pc and pD to be the probabilities that the sites of type А, В, C and D, respectively are occupied by Fe atoms. It is known from experimental evidence that the sites A and C are equivalent, i.e. Рд = Pß an<^ obviously

2 PA + PB + PC = 4 (1 " x) where x stands for the Al concentration.

We define now the order parameters a and ß as

PA = 1 - x + ax /За/

PD = 1 - x + ßx

/зь/

and from /2/ we have

PB = 1 - x - (2a + 3)x /3d/

Considering the occupation of the first neighbour sites we find that

sites В and D are surrounded by 8 A sites and the probability that n of these are occupied by Fe atoms is

p 8<n 'pA> - ( n ^ A (1-Ра>8'П /</

sites A are surrounded by 4 D and 4 В type first neighbours. The probabil­

ity that к of the D sites and 1 of the В sites are occupied by Fe can be expressed as

(7)

3

p4 (k' PD> • р4 а 'Рв> = ( k ) PD (1 ■ PD> k ( l ) pB (1_pB )4’)l /5/

The average magnetization per atom of the alloy is given by

11 - ПА PA + ПВ VB + nD PD /6/

where n* _ „ stand for the relative numbers of iron atoms at the given A , В , D

type of sites and obviously:

"A = °-5 .- V "в = °'25 ' PB' ■ nD = °-25 • PD

and У. „ r, stand for the average magnetic moments at the given type of

A f хэ f L)

sites, thus

4 4

5 j s I

'A

!

L *4р

4 <*'

p

d > p 4 (i"

p r

} * W k +Ä)

k=oi-=o

В Т е /7/

8 8

«В * PD = I p8 (n' Pa5 PF e (n) = I p 8 (n' PA> MD + P8 (4’ PA> VA /8/

n=o n=5

On substitution into /6/ we get

У - 0,5(1-x) (yA + yD ) - 0.5ox (yD - Уд) 19/

Let us look at some special cases.

1/ Complete disorder /g-phase/

a = ß = 1, thus рд = pB = pD = 1-x and therefore уд = yD and

У = (1 - x) yD .

2/ Perfect Fe^Al-type order

The excess or deficient A1 atoms due to the deviation from stoichiometry are responsible for the changes in the occupation of the D and В-type sites.

Two cases have to be distinguished .

(8)

4

а/ X — 0,25

Then а = ß = 1, thus рд = pQ and PB = 1 “ 4 x. By /7/ we have уD = yD and by /8/ уд = yD “ (4x)4 (yD - уд ) thus from /9/

у = (1 - x) yD - 0 .5(4x) (yD - уд ) .4

b/ x Ut. 0.25

Then рд = 1, Рв = 0 and by /2/ pß = 2 - 4x. By /7/ and by /8/ Уд = (2 - 4x)4 У thus

у = (0.5 - x) yD + 0.5 (2 - 4x)4 Уд •

3 / Perfect FeAl-type order

Then PA = 1, PB = PD = 1 " 2x, = D and 8

Уд = 1 P8 (n, PD ) yD + P8 (4, Pd ) уд , thus n=5

Я = (0.5 - x )уD + 0.5 уд .

D I S C U S S I O N

The solid line in Fig. 2 shows the average magnetization, as ob­

tained from equation /9/ for the order parameters evaluated from the dif­

fuse X-ray scattering data given in Fig. 3. The calculated values are not significantly sensitive to the order parameters, a 10 % change of the latter induces not more than about 2 % change in the former. The agree­

ment with the measured values which is satisfactory up to 25 at%. Al be­

comes gradually worse as the Al concentration increases.

The difference between the predicted and measured values in the alloys with more than 25 at% Al can be explained by the increasing dis­

order of the Fe^Al phase as ever more D sites are occupied at random by Al atoms and only the still remaining Fe^Al type clusters can contribute to the average magnetization of the alloy in the measure that

5 = ncl ücl , д о /

where nc ^ is the relative .number of Fe^Al type clusters, thus

(9)

5

ncl = 0.5 - p,

where p is the probability that a Fe atom in an А -site has 4 Fe neighbours in D-sites, which is given as

p = (2 - 4x)4 .

The average moment of such a "magnetic cluster" is

where yß is the average moment of Fe atoms at D-sites, i.e.

= (1 - 2x)ud .

On substitution into eq./lO/ the average magnetization produced by the randomly distributed Fe^Al type clusters is given as

у = 0.5 (2 ~4x)4 (yA + (1 - 2x)yD) /11/

The values calculated from eq./ll/ are shown in Fig. 2 by the broken line which gives a good agreement with the experimental data even in the critical range of concentrations.

The Fe moment уд , for 4 Al first neighbours, as evaluated from the reported average magnetization data is in good agreement with

yA = 1.8 + 0 . 1 yß obtained from Mössbauer data by using the expression

• w 3) - HF e (4> * a <“d - V + AH<

where Hp e (3) = 261 kG [7] and Hp e (4) = 210 kG [l2] are the iron hyper- fine fields measured at room temperature on alloys of about 25 at% Al for 3 and 4 Al first neighbours respectively; а(У0 ” ^д) is the change of the core polarization contribution, where a = 65 kG/Уц is the core po­

larization constant54 and Ah = 23 kG [e.g. 1~[ is the change in the con­

duction electron contribution due to the replacement of a Fe atom at D site by an Al atom.

a = 65 kG/yß was obtained from the hyperfine field of 145 kG corres­

ponding to Fe atoms without Al first; neighbours measured at about 50 atfc Al at 4°K [8, 13] , where the presence of a paramagnetic line shows that

the conduction electron contribution can be ignored.

(10)

6

The difference from уд = 1.5 + 0 . 1 uß that was determined from neutron diffraction on a Fe^Al specimen [9] can be explained by either a slight deviation from stoichiometry or non-perfect order of the Fe^Al specimen.

It is expected that the similar anomalous magnetic behaviour of the Fe-Si system can be explained in the same way.

A C K N O W L E D G E M E N T *

Thanks.are due to Professor L. Pál for many valuable discussions and for critically reading the manuscript and to Dr. C. Hargitai,

Dr. L. Cser and Dr. G. Grüner for the helpful comments.

*

(11)

L I T E R A T U R E

[1] A. Arrott and H. Sato: Phys. Rev., 114, 1420 /1959/

[2] W. Sucksmith: Proc. Roy. Soc., 171, 525 /1939/

[3] D. Parsons, W. Sucksmith and J.E. Thompson: Phil. Mag., 2' 1174 /1958/

[4] H. Danan and H.- Gengnagel: J. Appl. Phys., 39_, 678 /1968/

[5] R. Hergt, E. Wieser, H. Gengnagel and A. Gladun: Phys. Status Solidi, 41, 255 /1970/

[6] A. Lawley and R.W. Cahn: J. Phys. Chem. Solids, 20, 204 /1961/

[7] G.P. Huffman and R.M. Fisher: J. Appi. Phys., 3J3/ 735 /1967/

[8] G.P. Huffman: to be published in J. Appl. Phys.

[9] S.J. Pickart and R. Nathans: Phys. Rev., 123, 1163 /1961/

R.Nathans and S.J.Pickart: Magnetism III, 225 /1963/

[10] C.H. Cheng, K.P. Gupta, C.T. Wei and P.A. Beck: J. Physichem. Sol., 25, 759 /1964/

[11] T.M. Srinvasan, H. Claus, R. Viswanathan, D.J. Bardos and P.A. Beck:

Electronic Structure of Alloys of 3d Transition Metals with Aluminium [12] M.B. Stearns: Phys. Rev., 168, 588 /1968/

[13] G.K. Wertheim and J.H. Wernick: Acta Metall., 15/ 297 /1967/

(12)

- 8 -

• А ЭС OB ®D

Fe3 A l : • Э 0 Fe О Al Fe Al : • Э Fe О ® Al

gjg- Í

Site designations for Fe^Al and FeAl type superlattices

%

(13)

A

9

<

Fig.2 Average magnetization ofFe-Alalloysasafunction ofconcentration asextrapolated to0°K. AArrott andSato[lj•Sucksmith [2]□Parsonset.al

]

ОDananandGengnagel [4]

(14)

CA. S ЬЧ

- 10 -

-H№ Еч

СО

-lattice occupation asafunctionofcomposition /LawleyandCahn [6]/.TheAandCsitesareequivalent.

(15)
(16)

Kiadja a Központi Fizikai Kutató Intézet Felelős kiadó: Tompa Kálmán, a KFKI Szilárd­

testfizikai Tudományos Tanácsának elnöke Szakmai lektor: Vasvári Béla

Nyelvi lektor: Monori Jenoné

Példányszám: 255 Törzsszám: 71-5867

Készült a KFKI sokszorosító üzemében, Budapest 1971. augusztus hó

Hivatkozások

KAPCSOLÓDÓ DOKUMENTUMOK

From Table 4, it can be seen that at a given confining pressure, the peak deviator stress of the clay reinforced with untreated and treated coir fibres in- creases with increases

Increasing the hole current, at a certain voltage the inversion charge, and thus the field-strength in the insulator increases, which also increases the injected electron

iron ores the obtained slag has low viscosity at the metallurgi- cal temperature of the iron smelting.Furthermore, it is possible to decrease the melting point of the slag by

The rate of reaction increases as enzyme concentration increases (at constant substrate concentration). At higher enzyme concentrations, more enzymes are available to catalyze

In case a significant fraction of Rh atoms remained on the topmost layer of bimetallic clusters at room temperature at these coverages, and if there had been gold atoms in

Interestingly, clusters of microglia observed in the brain at sites of virus infection in P2Y12 −/− mice were located in the close vicinity of degenerated,

Labor cost increases the demand curve shifts inward at given net wage8. S,D

If the reduced inertia can be compared with the rotor inertia as it seen in Fig 6, regulation slows down (T increases) and the current load of the armature considerably