W wir . . monet muuten van de wedstemor .. 10 X 1. lut AL ..? : UNCLASSIFIED ORNL . . . " . : .. KI EXY 2. Senta 37. EVRAK ITU** O romo 428 (paper to be presented at IXth International Conference on Low Tempersture Proies, Columbus, Ohio, August 31 - September 4, 1964) ORNL - 428 ITIES 007 91964 م رکه - سیر S yperrine Structure Coupling in Ferric Ammonium Sulfate as a function of Magnetic field and Temperature F. :. Ovenshain, !. D. Roberts, C. F. Coleran, D. W. Forester Oak Fidge Tational Laboratory, Oak Ridge, Tennessee and J. O, whomson University o: Tennessee, Knoxville, Tennessee I: stüdyirs the hyperfine structure coupling in ferromagnetic or citiierronagrietic materiais by means or the Mössbauer effect the concept Oi 2. erective magnetic field bon has been widely used. A precise wurinicion oi !! may ve given in terms of the spin llwmilionian Hi cucciated with ti: ith magnetic ion of spin S; it: the magnetic solid: UIT leemi. .ini; + 5 J., S. , +382 5 (1) j=i bris fij C Here is is an externally applied magnetic field, I is the nuclear spin oi tho :" ion, the fourth term is the exchange interaction, the fifth term is the dipole-dipole interaction and the remaining symbols have their US usual reaning. In naterials such as metallic iron and Fe, , the exchange Research sponsored by the U. S. Atomic Energy Commission under contract ME - LEGAL NOTICE --- with the Union Carbide Corporation. The report m. peenrnd au u account of Doren.nl ornsond work. Neither Une Valls mau., noo the Commission, for any person ucun on bail of Croatian A, MIL MJ varrinty of ropesenulim, por or impiind, 10 roinnt um arru. racy, complement, or inelulnis of the lalornauon conuine In wis import, or that the mi of any information, maritus, methind, or procesu delom in wo report wy not infring prinulyomd right, or n. Aswne. may laMIIH. «10 respect an we va ol, or lor de macro rooting from the m of any information, order.hui, arthurd, er proces, decloud In we report. As im le the above, "prm kung an half of the cowalielon" includes any on. porn or contracur of the Connulou, of implor of me contractor, no tent that inplore or contractor of the Coa niosion, or apoyo a nuca contrac'or preparat, diomainte., provides accmi be, any Inor sulton pureut to do amoy on or contrat 0 Comiuolan, or to employmel nid oernali actor. -2- hüür.cüica is much larger than thie other terms in Hi, including in vichar the hyperfine structure coupiing. Correspondingly the spin- 5, relaxation time I will be very short compared with the nuclear 2000sion time I r when the temperature T is below the Curie temperaivre fonetic orderins efi'ects suppress the of? diagonal terms in AI;S;. i till ;-+ >>] the hyperfine interaction may then be described approximately 19 fs = 5nPne fiziert la) were cap = A <$iz> 16 43, +$426 = H149/372 ***/*T. To incore effective field model gives a good description of Mössbauer C.Vishnuments of the nis splitting occurring in metallic iron and I woübü be of interest to expiore magnetically dilute materials in which wie reiüive magnitudes of the terms in Hi may differ markedly from those cor le or 0203 Be substance of interest is the dilute parumagnetic material FC 1:50: 19.12!1,0. Here the exchange interaction is many orders of mag- z wcie smaller than in metallic iron, and the dipule-dipole interaction is rolig oc order or magnitude smaller. In this salt the dinolc-dipole izveracüion is thought to be much larcer than the exchance term in Hi, which now contains an additional term D(5,- 1/3 S(S + 1)). The coefficient D as riot iven measured, but may be of comparabic magnitude' to the di polizm ciäpoic orm. Vagnetic ordering in this material for 11 2 0 is presumably due to the D torn and dipolc-dipcle interaction and ling a Néel temperature ".shent 0.04**. Is one applies a magnetic field of about 10" oorsteds to tric salt ai temperatures in the iiquid helium ranga, a value for carnetic saturation may be produced. In the measurcnents described bilo: we have chosen to make a Szilabout one order of magnitude larger this tire siipole-iipole nná o terms which in turn are large compared to the oilier terms in 11. The magnetic ordering is now duc primurily to the ested field, in contxnst to metallic iron, where the spin ordering is Osü entirely due to the exchange interaction. Onc muy ühen inquire how um tho nas spectra observed in these measurements may be correlated with thic predictions of the effective fielú model. An additional interest in this measurement derives from the fact that ferrio alum has been used for man piy years in magnetic cooling studies. ás radioactive source wil3 prepared by diffusing about 70 mc of "Co into 0.0003 inch thick copper foil. The diancter of the active arcas was approximately 2.2 cm. The absorber was a polycrystalline sample of 54:2, 650,19.1211,0 containing 20 mg/cm- of natural Fc, pressed into the form SW. the same liquid helium bath, but the macnetic field was applied only to the übsorber. The direction of the magnetic field was normal to the planc or tio alisorber and parallel to the Y-ray propagation direction. Picasurements were made at H/T values of 14.10, 10.65, 8.0, 6.50, 5.72, 4.30, 2.86 ind We zero koc/ºk. A selection from these results is shown in Fig. l. 11 At the largest 11/m vnluc or 14.10 koe/ºk, Fix, la, a hyperfire structiire op:0c truri of f'our lines was obverved in which the relative Bpac- incu oi' lines corresponded to those observed in ferromagnetic iron, and ind i the lineu showed widtho 10.05 mm/sec. me line width found for .càü source at W.2°x using a Xilc(08)6:311,0 absorber at 300°K was 0.47 mm/sec. me rehatively smali line brondering in this spectrum compared with that 51 ( Co TO vidad for the smaller 11/" values discussed below results from the fact wiūt liere 2 SM is near cuturation, i.c., the electron spins spend most 0:' the time in the energetically most favored orientation S, E. 5/2. For O v ürini single crystal or the onlt one would expect the lines to approach üeir ilüural width as 11/" approacics infinity. Assuming for the xresent ühjat I, is opposite in sien to the applied lielu II, and using Eqs. (2-4) é to extrapolato to saturation magnetization, the observed splitting corresponds 101... = 991 20 koe at < 5,> /SE 1. This may be compared with the vallie 552 10 koc measured at 1.2°% for antiferromagnetic reg.g, in which tric iron utoms are also in the i'erric state. mici! 2 3 p The other spectru, or which i'ie. 10 is a typical example, show a very marked procressive asynmetric brondening and decrense of ampiitude of the outer lines us !1/7 decreases. In sharp contrast with this, within statisti- IS SOS cal error the amplitudes and widths of the inner lines are very nearly in- deperdicat 08 !1/over the range from 14.10 to 2.86 koe/ºk. The arrows in .. i show where the peaks of the outer lines would appear if their positions con lowesi 993. (2-). Tie separations of the outer line pair and those of the inner pair for üll of our data are shown as a function of B11/kt in Fig. 2. mie solid curves - 5 are a plot of <1,> given by Eq. (") normalized in each case to the splitting of the corresponding line pair at the largest value of H/T. Tuc spacing of the inner pair appears to follow the Brillouin function CG. () within the statistical uncertainties of the measurements, but me Sureme the peak separation of the outer lines clearly falls above the calculated curve as 11/T decreases. In discussing these measurements, three characteristic times are of C @ importance. These are the nuclear life time are the spin-spiti relaxation time 2a and the nuclear precession time 7. Clearly T / Ty must be small compared with one in order to have a resolved spectrum and TIT must be large compared vith one so that fluctuations of the electron spin riacnetization may have a negligible eifect on the line widths. As mentioned avove, these conditions are well satisfied in metallic iron. In contrast, Cr17, in ferric ammonium alum is much smaller and possibly comparable with unity. llere fluctuation effects may be expected to have a strong in- fluence on the observed Mössbauer spectrum. In this context one may note that the broadening effects are much more pronounced for the outer lines, for which /(m 4/I. - m4/I) I is large, than for the inner lines, and that is the outer lines become broader they also become progressively more asymmetric. This asymmetry can be attributed to the selection rules correspond- ing to the dipole-dipole and D terms which give rise to the electronic spin- spin relaxation process. These selection rules are such that spins in energetically highly unfavored orientations, which will have very few neighbors suitably oriented for spin exchange, will relax unusually slowly towards more probable orientations. Another possible contribution to the asymmetry may come from én, = 0 transitions. Because of the presence of off-diagonal OXTC TON Il terus in the spin Hamiltonian, these transitions will not be completely for- bidder. he is spectrum of ferric ammonium alum was also observed under field- free cordiitions at temperatures of 1.7, 2.25, 4.2, and 300°. In each case Q res the spectrum showed a sincle rather broad linę, the width at 4.2°K being 1.3 m/sec. This is to be contrasted with the resolved multi-line spectra r observer under field-free conditions in magnetically very dilute materials by hertheim et al. It is clear that under the conditions of this experiment the effective s'iclü model does not give an adequate description of the behavior of f'erric üonium alum. The discrepancies are quantitative, in its failure to account . for the positions of the outer peaks, and qualitative, in that the model assumes that the lines will show only the natural wirith associated with the ili clear lifetime. Both types of discrepancy may well be removed by a theory which takes account of fluctuations in the electron spin magnetization. -7- REFERENCES 1. Visiting scientist from the Atomic Energy Research Establishment, arvell, England. Oak Ridge Graduate Fellow from the University of Tennessee, Knoxville, Tennessee. York (1962). H. 3. S. Preston, S. S. llanna and J. Heberle, Phys. Rev. 128, 2207 (1962). 5. %. Ono, Y. Ishikawa, A. Ito, and E. Hirahara, J. Phys. Soc., Japan, Suppl. B-I, 17, 129 (1962). 6. m. 7. Cooke, Proc. Phys. Soc. (London) A62, 269 (1949). 8. G. E. Wertheim and J. P. Remeika, Bull. Am, Phys. Soc. 2, 464 (1964). FIGURE CAPTIONS rico? - Pyperfine structure spectra of 'l'e in ferric ammonium alum a) 11.= 24.0 koe, T = 1.70K b) 1 = 24.0 koe, T = 4,2°K Fig. 2 - Peas separations in nyperfine structure spectrum of ferric ammonium alum as a function of BH/KT. Solid circles - separation of outer peaks Open circles - separation of inner peaks For explanation of curves see text. Curves S UNCLASSIFIED ORNL-DWG 64-3044 R so H = 24.0 koe T = 1.7 °K runkát, ki - war, wa ARBITRARY UNITS (0) Castindo H = 24.0 koe r = 4.2°K de ARBITRARY UNITS *** 14 12 10 8 8 10 12 14 6 4 2 0 2 4 6 RELATIVE VELOCITY (mm/sec) Iron-Ammonium Alum. Figure 1 Avo (mm/sec) 20 0.25 0.50 a=0.67x104 HT 0.75 Figure ? 1.00 1.25 ; ORN-OVG (:--0145 UNCLASSIFIED) 4.50 Avi (mm/sec) . - - - - . - . .. - - . .. ... -- - . . .-. . . . . . . . . . :o. w m .- -. ~_ arri v a t * t **. A . . S-am a* - - - B ': .. DATE FILMED 121/ 41 164 AL . . - Sa 3 . .. * * - . -LEGAL NOTICE 22: LEX * * - SS This report was prepared as an account of Government 3ponsored work. Neither the United Sintos, nor the Commission, nor any person acting on behall of the Commission: A. Makes any warranty or representation, expressed or implied, with respoct to the accu- racy, complctoness, or unofulness of tho information contained in this report, or that the use of any information, apparatus, method, or procons disclosed in this roport may not infringe privately owned rights; or B. Assumos any liabilities with respect to the use of, or for damages resulting from the use of any information, apparatus, method, or proco88 disclosed in this reporl. As used in the abovo, "person acting on behall of the Commission" includes any em- ployce or contractor of the Commission, or omployce of such contractor, to the oxtent that such employoe or contractor of the Commission, or omployee of such cor.tractor properes, didacminntos, or provides access to, any information pursuant to his employment or contract with the Commission, or his employment with such contractor. . & . * 2 - 2 . ES N' . . . ET Ws ET TRR 2 ai . END ht 777 >