• Aucun résultat trouvé

STRESS-INDUCED MAGNETIZATION

N/A
N/A
Protected

Academic year: 2021

Partager "STRESS-INDUCED MAGNETIZATION"

Copied!
4
0
0

Texte intégral

(1)

HAL Id: jpa-00214066

https://hal.archives-ouvertes.fr/jpa-00214066

Submitted on 1 Jan 1971

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.

STRESS-INDUCED MAGNETIZATION

D. Craik, R. Fairholme

To cite this version:

D. Craik, R. Fairholme. STRESS-INDUCED MAGNETIZATION. Journal de Physique Colloques,

1971, 32 (C1), pp.C1-681-C1-683. �10.1051/jphyscol:19711238�. �jpa-00214066�

(2)

JOURNAL DE PHYSIQUE Colloque C 1, supplbment au no 2-3, Tome 32, Fbvrier-Mars 1971, page C 1 - 681

STRE S S-INDUCED MAGNETIZATION

D. J. CRAIK and R. J. FAIRHOLME (*)

University, Nottingham (*) Novo a t Plessey Research Laboratories, Caswell, England

RbsumB. - Les courbes d'aimantation de forme complexe sont obtenues en appliquant des contraintes des ferro- magnktiques soumis a de faibles champs de pulsation. Un ensemble consbquent de donnkes est dkcrit en termes de para- metres-empiriques et de courbes caractkristiques rendant compte des rotations des mouvements irrkversibles de parois et de transformations compl~tes des domaines. L'observation des domaines et des cycles de contraintes permettent les interprktations.

Abstract. - M-a curves of complex form are obtained on applying stresses to ferromagnetics in small bias fields.

A large amount of data is described in terms of empirical parameters and characteristic curves, the explanations of which involve rotations, irreversible wall motion and complete transformations of the domains. Domain observations and stress cycling aid the interpretations.

1. Introduction. - The application of tensile or compressive stresses to ferromagnetic materials in small bias fields can cause large changes in magnetiza- tion. The bias fields referred to are generally such as t o induce only a low level of magnetization compared with that induced by the subsequent application of the stress, but no consistent stress-induced changes are observed in the complete absence of a bias field. The coupling between the stress and magnetization is magnetostrictive and the stresses can be equated to effective fields using the magnetostriction coefficients of the materials concerned. Craik and Wood [l]

illustrated the complexity of the magnetization-stress (M-o) curves for substantial stresses still below the elastic limit, and gave references to earlier work. In the present paper an attempt is made to classify the types of M-o curve which may be obtained for a wide range of materials, and empirical parameters are introduced to aid comparison of the response of different materials.

The effects of repeated stress cycling also help to elucidate the underlying mechanisms.

2. Method of measurement. - The important fea- tures of the measurements are that they are made on thin strips, with low demagnetizing factors, fitting into

Material - Ni

85/15 NiFe 8211 8 NiFe 80120 NiFe 70130 NiFe 50150 NiFe Mild steel

Random 3 % SiFe Cube 3 % SiFe Single 3 % SiFe Mumetal Hard steel Cobalt

M-0 sequence

-

? (9 (iii) (iv) (iv) - (iv) (iv) (iv) (iv) -

?

a yoke and with compensation for residual demagne- tizing fields. This avoids variable demagnetizing effects but involves the necessity for high sensitivity measure- ments of the magnetization changes. A voltage-to- frequency converter with effective microvolt sensitivity was used to integrate the voltages induced in a coil around the specimen, giving AB = 1 gauss or AM = 0.08 e. m. u. detectable for an iron specimen with 0.05 cm2 section. Specimens were ac demagne- tized before applying the bias field and stresses. The earth's field was annulled.

3. Classification of M-o curves. - On applying a stress the magnetization rises and may approach a steady value or even pass through a maximum and fall once more. On removing the stress M may rise further or fall.

Behaviour of the two most extreme types can be conveniently specified in terms of the magnetization induced by the bias field alone, M,, the stress-induced remanent magnetization Mtr or Mc, for tension and compression respectively, the maximum induced magnetization with o increasing, Mtm or Mcm, and the stress otm or o,, at which this occurs. This data is outlined in table 1 for all the specimens measured.

Compression

M-0 sequence

-

(iv) (iv) (iv)

?

(9 (i) (iii) (iii) (iii) (iii) (i) ?

Tension

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19711238

(3)

C 1 - 682 D. J. CRAIK, R. J. FAIRHOLME Description is complicated by the dependence of the

form of the results on the peak value of the total effec- tive field (bias field and equivalent stress field) attai- ned. Four sequences appear t o arise for various mate- rials, in terms of increasing fields (Fig. 1). All thevarious

FIG. 1. - Characteristic sequences of M-s curves for differing peak stresses. The origin in these is the magnetization induced by the bias field alone. Table 1 shows which materials exhibit

each type or sequence of curves.

types of M - o curves are incorporated in these sequences and the sequence applicable to each material shown in table 1. The first two sequences are characterised by an approach to a limiting value of M, and indeed (i) is very similar to a family of magnetization-field curves. (iii) and (iv) are jointly distinguished by the onset of a peak value of M at high values of o and distinguished from each other by the shape of the curves with decreasing a and the magnitude of the dip in (iv) which may extend to the demagnetized state.

(ii) might be considered as a special case of (iv) : for materials giving this type of result no value of o will produce a maximum M with subsequent decrease, at low bias fields, but increasing the bias field may give results characterised by (iv).

Sequence (i) applies to materials where simple considerations of the magneto-elastic and anisotropy torques suggest that the results can be regarded as indicating large rotations in an effective field H, = 3 Aa/2 Ms. This corresponds well with the general appearance of the curves, with irreversibility due to rotational hysteresis.

Sequences (iii) and (iv) apply to materials with negli- gible rotation (lo 4 K,). It was natural to associate this behaviour with movement of domain walls.

Sequence (iii) may be explained using an extended form of Brown's theory [l, 21 with the smoothing effect of 1800 walls. In sequence (iv) the initial increase in the magnetization is followed by a complete break- down in the domain structure as observed by Bates and Carey 13, 41. Using the longitudinal Kerr effect it

was possible to confirm (Fig. 2)Xthat domains magne- tized normal to the strip axis (i. e. the direction in which the bias field and stress were applied) gradually replaced those parallel to the field axistas the turnover and decrease in M occurred. Clearly, with all 1800 domains normal to the strip axis the measured magne- tization must be zero.

It is of considerable interest to note that while our observations of the domain rearrangement, and those of Bates and Carey, were made on singly-oriented and

FIG. 2. -Domains in SiFe (Kerr effect) related to the M-a curve.

cube-textured silicon iron respectively this rearrange- ment must now be considered a very common pheno- menon (in mild steel, nickel and some NiFe alloys) unless an alternative explanation of the type (iv) sequence can be put forward.

4. Stress cycling. - A basic feature of the mecha- nism, which leads to the stress-induced remanence,

Q 4 t . P 40

a o r

complete cycles

b)

S t r e s s c y c l e s otter

snhys t e r e t i r d e m ~ ~ n e t i s a t i o n

6 (%/mme'

-4

-9. 0 s,

4

a) ~ n l t i a l s t r e s s cycLes

FIG. 3. - ( a ) M-a cycles for random SiFe (3 %) after ac demagnetization and application of bias field.

(b) M-a cycles after anhysteretic magnetization in a bias field.

(4)

STRESS-INDUCED MAGNETIZATIOPU' C 1 - 6 8 3 is the irreversibility of the motion of walls under the

influence of stresses, corresponding to familiar hyste- resis in applied field. Knowing that the cycling of fields can lead to a relatively reversible state, it is expected that the repeated application of stresses should have some corresponding effect.

Three specimens were chosen : random SiFe (low rotation), 70130 NiFe (large rotation) and hard steel (small changes corresponding to Brown's theory). The specimens were ac demagnetized, exposed to the bias field, and stress cycled commencing always with tension for reproducibility. In each case large increases in magnetization were followed, after a few cycles, by the description of reproducible loops about a steady mean magnetization (Fig. 3a). The reproducible loops were obtained almost immediately for the 70130 NiFe with only a small increase in the mean magnetization.

This behaviour re-inforces the foreging arguments.

Rotation, in the 70-30 NiFe, is very nearly reversible.

In the iron specimens the ac demagnetization brings the domain walls to an equilibrium configuration, but in a bias field this same configuration, stabilised by impediments to wall motion, no longer corresponds

to equilibrium. The cycled stresses move the walls until, for zero stress, they approach a new configura- tion which does correspond to equilibrium in the bias field. Now the effect of an applied stress is much smaller and much more nearly reversible since the walls are moving around equilibrium and not approach- ing equilibrium.

Two further obvious studies correlate with this.

First, cycled magnetic fields are found to induce the same resultant magnetization as the cycled stresses, i. e. the stress cycling leads to the familiar anhysteretic state for a given bias field. Secondly, the initial M-a cycles around the anhysteretic point are very similar to those after repeated stress cycling (Fig. 3b).

5. Summary. - Different types of M-a curves can be explained by either rotations, irreversible wall motion in effective fields corresponding to the stresses, or complete domain rearrangements. Stress cycling gives results which help to discriminate between rota- tion and wall motion and, particularly, illustrate the association of large remanent magnetizations with the irreversibility of domain wall motion.

References

[I] CRAIK (D. J.), WOOD (M. J.), J. Phys., D, 1970,3, 1009.

[2] BROWN (W. F.), Phys. Rev., 1949, 75, 147.

[3] BATES (L. F.), CAREY (R.), PYOC. Phys. Soc., 1960, 76. 754.

[4] CAREY (R.), PYOC. Phys. Soc., 1962, 80, 934.

Références

Documents relatifs

Up to now, we have seen that the main effect of the Oersted field is to tilt the NiFe magnetization in the direction transverse to the nanowire. the tilt angle) undergoes

The exposure of iron films to about 1 t o 10 L of hy- drogen leads to a decrease of the magnetization equiva- lent to a reduction of the surface moment of the order of 0.5

Figure IV.9: [111, 196]: Examples of SAW-FMR curves for (Ga,Mn)As and (Ga,Mn)(As,P) evidencing dierent anisotropies: for an out-of-plane magnetized layer, the eld must be

Strain differences in hypothalamic-pituitary-adrenal (HPA) axis cross- sensitization were determined by plasma corticosterone (CORT) levels and hippocampal gene expression

Conversely, the application of a compressive stress tends to promote the growth of closure domains at the expense of the primary domains, which means that now the 900

NEUTRON DEPOLARISATION AS A METHOD TO DETERMINE THE MAG- NETIZATION, THE MEAN DOMAIN SIZE AND THE MEAN SQUARE COMPONENTS OF THE INNER MAGNETIZATION OF

Normalized MOKE intensity (thus giving normalized magnetization) curves (dots) versus magnetic field applied either parallel or perpendicular to the steps measured in the 10 ◦

The response of any magnetic field sensor, which depends upon the magnetization direction of its sensing elements, is subject to intrinsic magnetic noise arising